G Model ARTICLE IN PRESS - appliedminex.com · skarn deposit Sangan Iran a b s t r a c t The Sangan...

C

MI

FKa

b

c

d

e

f

a

ARRA

KGCMISI

1

smtanaSDea

o

h0

ARTICLE IN PRESSG ModelHEMER-25446; No. of Pages 21

Chemie der Erde xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Chemie der Erde

journa l homepage: www.e lsev ier .de /chemer

ineral geochemistry of the Sangan skarn deposit, NE Iran:mplication for the evolution of hydrothermal fluid

atemeh Sepidbar a,b, Hassan Mirnejad a,c,∗, Jian-Wei Li b, Chunjing Wei d, Luke L. George e,ingsley Burlinson f

Department of Geology, Faculty of Sciences, University of Tehran, Tehran 14155-64155, IranState Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Wuhan 430047, People’s Republic of ChinaDepartment of Geology and Environmental Earth Sciences, Miami University, OH 45056, USAState Key Laboratory of Orogenic Belts and Crustal Evolution, Peking University, Beijing 100871, People’s Republic of ChinaSchool of Physical Science, University of Adelaide, Adelaide, SA 5005, AustraliaBurlinson Geochemical Services Pty. Ltd., Darwin, NT, Australia

r t i c l e i n f o

rticle history:eceived 18 April 2017eceived in revised form 5 July 2017ccepted 14 July 2017

eywords:arnetlinopyroxeneagnetite

ron skarn deposit

a b s t r a c t

The Sangan iron skarn deposit is located in the Sabzevar-Dorouneh Magmatic Belt of northeastern Iran.The skarn contains zoned garnet, clinopyroxene and magnetite. Cores and rims of zoned garnets aregenerally homogeneous, having a relatively high �REE, low �LREE/�HREE ratios, and positive Eu anoma-lies. The cores of the zoned clinopyroxenes are exceptionally HREE-rich, with relatively high �REEand HREE/LREE ratios, as well as positive Eu anomalies. Clinopyroxene rims are LREE-rich, with rela-tively low �REE contents and HREE/LREE ratios, and do not have Eu anomalies. Magnetite grains areenriched in LREEs in comparison with the HREEs and lack Eu anomalies. Variations of fluid composi-tion and physicochemical conditions rather than YAG-type substitution mechanism are considered tohave major control on incorporating trace elements, including REE, into the skarn mineral assemblage.

anganran

Based on baro-acoustic decrepitation analysis, the calc-silicate and magnetite dominant stages wereformed at similar temperatures, around 350–400 ◦C. In the Sangan skarns, hydrothermal fluids shiftedfrom near-neutral pH, reduced conditions with relatively high �REE, low LREE/HREE ratios, and U-richcharacteristics towards acidic, oxidized conditions with relatively low �REE, high LREE/HREE ratios, andU-poor characteristics.

Crown Copyright © 2017 Published by Elsevier GmbH. All rights reserved.

. Introduction

Skarn deposits are genetically closely related to magmatic intru-ions. They form as hydrothermal fluids released at the end ofagmatic crystallization and interact with carbonate country rocks

hrough metasomatism (Meinert and Nicolescu, 2005). Mineralssemblages often vary between different skarn zones, but gar-et, clinopyroxene, and magnetite are normally the most commonnd widespread minerals in skarns (e.g., Jamtveit et al., 1993;omarin, 2004; Ciobanu and Cook, 2004; Chang and Meinert, 2004;

Please cite this article in press as: Sepidbar, F., et al., Mineral geocheevolution of hydrothermal fluid. Chemie Erde - Geochemistry (2017),

ziggel et al., 2009; Ismail et al., 2014; Zhao and Zhou, 2015; Xut al., 2016). Garnets from plutonic rocks are generally Fe-richnd exhibit a weak increase in Mn content from core to rim (e.g.

∗ Corresponding author at: Department of Geology, Faculty of Sciences, Universityf Tehran, Tehran 14155-64155, Iran.

E-mail address: [email protected] (F. Sepidbar).

ttp://dx.doi.org/10.1016/j.chemer.2017.07.008009-2819/Crown Copyright © 2017 Published by Elsevier GmbH. All rights reserved.

Mirnejad et al., 2008; Samadi et al., 2014; Ismail et al., 2014; Xuet al., 2016), whereas garnets originating from skarns are oftencharacterized by Al-rich cores and Fe3+-rich rims (Gaspar et al.,2008). Trace elements in skarn minerals are useful geochemicaltracers (Cheng et al., 2012; Schmidt et al., 2011). For example, mag-netite from genetically different types of deposit displays varioustrace element patterns depending on fluid/water/melt composi-tions and processes of magnetite precipitation (e.g., Frietsch andPerdahl, 1995; Dare et al., 2014). Thus magnetite compositionmay reveal the physicochemical conditions under which it wasformed. During contact metamorphism and hydrothermal alter-ation of carbonate-bearing rocks, garnet and pyroxene grow undera variety of physicochemical conditions (Fernando et al., 2003; Kim,2006; Martin et al., 2011). These minerals commonly display com-plex chemical zoning patterns that may record the physicochemical

mistry of the Sangan skarn deposit, NE Iran: Implication for the http://dx.doi.org/10.1016/j.chemer.2017.07.008

evolution of the hydrothermal system. Chemical composition ofgarnet and pyroxene has been used to solve various geologicalproblems including understanding the physicochemical condi-

dx.doi.org/10.1016/j.chemer.2017.07.008


http://www.sciencedirect.com/science/journal/00092819

http://www.elsevier.de/chemer

mailto:[email protected]


ING ModelC

2 e der E

taih2iaotimog2dmtb

patmdabwtGart(sdlItp

2

so(ibtsml3tc(

bpdaacSz

ARTICLEHEMER-25446; No. of Pages 21

F. Sepidbar et al. / Chemi

ions of formation of both skarn minerals and rocks (Chakrabortynd Ganguly, 1991); fO2 changes in skarn (Kim, 2006), record-ng regional metamorphic history (Whitney, 1996) and reflectingydrothermal skarn fluid evolution (Gaspar et al., 2008; Smith et al.,004; Somarin, 2004; Ismail et al., 2014; Xu et al., 2016). Fluid

nclusion studies are mainly carried out on transparent miner-ls, despite the fact that economic minerals of interest are usuallypaque. To understand the fluid conditions during deposition ofhese deposits it is necessary to study opaque minerals. However, its difficult but there are three methods to study inclusions in opaque

inerals as the baro-acoustic decrepitation method, microscopebservations using infrared light, or mass spectrometric analyses ofases released during decrepitation of the sample (Burlinson, 1984,012). Magnetite is not transparent under IR, so the baro-acousticecrepitation method was used to study the fluids in this opaqueineral as well as garnet grains as transparent mineral. In addition

o estimating the formation temperature of the opaque minerals,aro-acoustic decrepitation detects the presence of CO2 rich fluids.

The Sangan iron skarn deposit, northeastern Iran (having aroven reserve of over 1000 Mt iron ore @ 53% Fe, Sangan represents

world-class iron skarn deposit), is located at the eastern edge ofhe Sabzevar-Dorouneh magmatic belt (SDMB) in the Sangan Mag-

atic Complex (SMC) (Fig. 1a). The skarn and the iron orebodiesefined by surface mapping and diamond drilling extend over 8 kmlong an E-W trending zone and occur at depths down to 600 melow the surface (Golmohammadi et al., 2015). Previous studiesere mainly limited to the alteration and mineralization charac-

eristics of selected parts of the Sangan deposit (Karimpour, 2004;olmohammadi et al., 2015), thus the fluid evolution of the depositnd chemical compositions of garnet, pyroxene and magnetite,emains incompletely understood. In this study, we report new tex-ural characteristics, systematic major, trace and rare earth elementREE) contents of garnet, pyroxene and magnetite and fluid inclu-ion studies in garnet and magnetite from the Sangan iron skarneposit. Data were obtained by electron microprobe (EMPA), in-situ

aser ablation inductively coupled plasma-mass spectroscopy (LA-CP-MS), and baro-acoustic decrepitation. The aim of this study waso decipher the fluid evolution responsible for crystallizing garnet,yroxene and magnetite in Sangan skarn.

. Geological setting

The oldest unit in the Sangan area is a thick pile (∼1500 m) ofedimentary rocks, consisting of Lower Jurassic shale and siltstone,verlain by Upper Jurassic to Cretaceous limestone and dolomiteKermani and Forster, 1991) (Fig. 1b). This sedimentary sequences intruded by granitoid intrusions or is unconformably coveredy volcanic rocks (Fig. 1b). Granitoids are extensively present inhe north of the Sangan area and exhibit a weak alteration con-isting of sericite and chlorite. They are pinkish grey and displayedium to coarse-grained granular and porphyritic textures. Pub-

ished zircon U-Pb ages of the Sangan intrusions range from 42 to8 Ma (Golmohammadi et al., 2015). At the contact between grani-oids and carbonate rocks, there is a thermal metamorphic aureoleontaining iron skarn mineralization (Golmohammadi et al., 2015)Fig. 1b).

The calc-silicate-dominant stage at Sangan is characterizedy replacement of limestone by andradite garnet, hedenbergite,lagioclase and lesser epidote, and rare biotite. The magnetite-ominant stage overprints calc-silicate-dominant stage mineralsnd produced amphibole, calcite, chlorite, phlogopite, magnetite


nd sulfides. Iron mineralization and associated skarn at Sanganonsists of seven orebodies from west to east (Fig. 1a): A’, A, B, C-outh (Cs), C-North (Cn), Baghak (BA), and Dardvey (D). Two garnetones, namely the garnet skarn and the garnet-clinopyroxene skarn

PRESSrde xxx (2017) xxx–xxx

zones, formed during the calc-silicate dominant stage. Garnet skarnzones are abundant in the northern part in the areas A’, A, Cn, Csand D (Fig. 1). Field observations indicate that the large magnetitebodies, commonly associated with these skarns, are located alongfaults or fracture zones. In the Cn skarn, garnet is well represented,forming a zone up to 30 m wide, while the garnet-pyroxene skarnforms a narrow zone, only a few cm wide, similar to the garnet-clinopyroxene zone in D skarn. The contact between the graniteand garnet skarn, where observed, is sharp and garnet skarn gener-ally grades outwards into garnet-clinopyroxene skarn. Associatedmagnetite bodies from the magnetite dominant stage are presentin orebodies A, B, Cn and Cs.

3. Analytical methods

3.1. Major-oxide element chemistry

Ten representative samples from orebodies A, A’, B, Cn, Csand D were chosen to investigate the origin and evolution ofhydrothermal zoned garnet, pyroxene and magnetite in the deposit.Following detailed textural and petrographic studies, twenty-twopolished sections of representative garnet, pyroxene and magnetitewere selected for systematic major and trace element analyses.Electron microprobe analysis (EMPA) was performed to acquirecompositional data. Mineral compositions were determined at theState Key Laboratory of Geological Processes and Mineral Resources(GPMR), China University of Geosciences (CUG), Wuhan, usinga JEOL JXA-8100 EMPA Analyzer equipped with a wavelength-dispersive spectrometer (WDS). Analytical conditions were: anaccelerating voltage of 15 kV, a beam current of 20 nA and a 2 �mfocused electron beam. Natural and synthetic mineral standardswere used for calibration, and matrix correction was done usingthe ZAF method to convert maps of X-ray intensity data into two-dimensional maps of oxide wt% and cation proportions for the eightmajor rock-forming elements. Element peaks and backgroundswere measured for all elements with counting times of 10 s and5 s (except for Ti and Mn, for which count times were 20 s and 10 s,respectively). The accuracy of the reported analytical values is 1–5%relative depending on the element abundance.

3.2. Trace and rare earth element chemistry

Trace element analysis of garnet, pyroxene and magnetite wasconducted at the Key Laboratory of Orogenic Belts and Crustal Evo-lution, Peking University, using an Agilent 7500c ICP-MS systemconnected with a 193 nm ArF Excimer laser system (COMPexPro102) with automatic positioning system. Ablation spot size was60 �m with a repetition rate of 5 Hz and laser energy of 100 mJ.Helium was used as the cell gas. The acquisition times for thebackground and the ablation interval were 15–20 s and 45–60 s,respectively. Synthetic glasses NIST 610, 612, and 614 were usedas external standards. The calculation of trace element concentra-tions was performed by GLITTER version 4.4.2. The accuracy of thereported analytical values is less than 5% relative depending on theabundance of the element.

3.3. Baro-acoustic decrepitation of fluid inclusions

Baro-acoustic decrepitation analysis was performed to deter-mine the fluid composition and formation temperature of trappedfluid in two magnetite samples (B-4 and C-4) from high-grade mas-sive magnetite ores from B and C orebodies and in one garnet


(A-1) hand sample of garnetite from A orebody at the laboratoryof Burlinson Geochemical Services Pty. Ltd. in Darwin, Australia.The analyses were done on 0.5 g of sample. There was minor car-bonate and sulfide in B-4 and C-4, respectively. Sample B-4 was


ARTICLE IN PRESSG ModelCHEMER-25446; No. of Pages 21

F. Sepidbar et al. / Chemie der Erde xxx (2017) xxx–xxx 3

Fig. 1. (a) Map of Iran and geological map of the Sangan region including location of the Sangan deposit and the Eocene plutonic and volcanic rocks based on 1:250,000geological maps of Taybad (Alavi Naini, 1982). Locations of A, A’, Cn, Cs, D and Ba skarn types are shown; (b) Stratigraphic columnar section of the Sangan area (modifiedafter Mazaheri, 1996).

Fig. 2. Representative hand specimens of garnets from different types of Sangan Fe skarn. (a) Red to brown, fine-grained euhedral garnet crystals from A skarn; (b) Greent dral gF d browi

wstarsa2wsa(cp

o gray, fine-grained subhedral garnet crystals from A’ skarn; (c) Coarse dodecaheine-grained garnet and associated calcite from Cs skarn; (e) Fine to medium grainen this figure legend, the reader is referred to the web version of this article.)

ashed with 10% HCl to remove carbonate while the sulfide inample C-4 was separated magnetically and by hand. After prepara-ion, magnetite sample B-4 consisted of >95% magnetite/haematitend was weakly magnetic, while magnetite sample C-4 containedare pyrite grains. Most sulfide was eliminated during magneticeparation, but minor mineral contamination remained. Prior tonalysis the sample was crushed and sieved into a fine fraction of00–420 �m (80–40 mesh). The prepared samples were analyzedith the BGS model 105 decrepitometer (Burlinson, 1990). The

amples were heated at 20 ◦C per minute over a range of 100–800 ◦C


nd fluid inclusion bursts were simultaneously recorded digitallyBurlinson, 1988). The decrepitometer was calibrated for qualityontrol of the results using standard samples. Decrepigrams werelotted using Gnuplot, version 4.6 (http://www.gnuplot.info). The

arnet grains of red to brown color from Cn skarn that is replaced by calcite; (d)n-colored garnet grains from D skarn. (For interpretation of the references to color

ore formation temperature was estimated based on the “onset tem-perature” (temperature at the toe of the peak), which marks thestart of massive decrepitation of fluid inclusions.

4. Petrography

The Sangan deposit is a calcic and magnesian type skarn formeddue to the interaction of magmatically-derived hydrothermal fluidsand two different protoliths; limestone in the west, and dolo-stone in the east (Mazaheri, 1996). The calcic skarn in the western


zone contains andradite (And)-grossular (Gro)-rich garnet, Ca-Fe-rich pyroxene, ferrohastingsite associated with wollastonite,plagioclase, and K-feldspars in the endoskarns, and ferroactino-lite, Fe-rich chlorite, clinochlore, calcite, and magnetite in the


http://www.gnuplot.info





4 F. Sepidbar et al. / Chemie der Erde xxx (2017) xxx–xxx

Fig. 3. Representative optical characteristics of garnets from different types of Sangan Fe skarn. (a) Photomicrographs of garnet from A skarn under crossed polarizedlight = XPL; (b) Dodecahedral garnet and polysynthetic twinning texture in garnet from A’ skarn; (c) fracture filled with calcite in garnet from Cn skarn; (d) massive magnetitg d phlos clinop

efM2(Sscs(

(tofiaaMgame

ra(geggaacaemi

rains and associated pyrite from calcic skarn; (e) massive magnetite and associatekarn;. Grt = garnet; Cc = calcite; Mt = magnetite; Phl = phlogopite; Py = pyrite; Cpx =

xoskarns. The magnesian skarn in the eastern zone containsorsterite, diopsidic pyroxene, phlogopite, actinolite, serpentine,

g-chlorite clinochlore, talc, and magnetite (Golmohammadi et al.,015). Garnet is the predominant mineral in both skarn zonesgarnet and garnet-clinopyroxene skarns) in the Sangan deposit.mall (0.5–1 cm) euhedral garnet crystals are very common in allkarns (Fig. 2a–d) while medium to large (1–2 cm) euhedral garnetrystals are located in Cn orebody (Fig. 2e). Garnet-clinopyroxenekarn, cropping out as greenish brown medium to fine-grained0.10–0.25 cm) rocks, is dominant in the A’ skarn (Fig. 2b).

Large garnet grains from Cn often exhibit replacement by calciteFig. 2c). The garnet skarns typically preserve granoblastic tex-ures consisting of zoned garnet with minor and variable amountsf calcite, quartz, biotite, amphibole, chlorite, magnetite, and sul-des. Due to physical and chemical condition variations, puritynd perfection of garnet crystals during growth conditions (Hariyand Kimura, 1978), this mineral is weakly anisotropic (Fig. 3).agnetite, the most abundant oxide mineral associated with the

arnet skarn, commonly occurs as medium to large grains (Fig. 3a),lthough magnetite inclusions within garnet are common. Theagnetite grains are anhedral to subhedral (20–300 �m) and are

nclosed by pyrite in some locations.Garnet-clinopyroxene skarns typically contain garnet, clinopy-

oxene, calcite and quartz, together with K-feldspar ± plagioclasend magnetite. Generally, clinopyroxenes form fine-grained0.1–0.2 cm) anhedral crystals or disseminated grains along garnetrain boundaries. Subhedral to euhedral, pale green clinopyrox-ne (0.1–0.2 cm), in addition to occurring as inclusions in garnetrains, forms large granoblastic or granular intergrowths witharnet (Fig. 3b). Garnets are normally fresh, but in some placesre partly replaced by an assemblage consisting of calcite, quartznd magnetite ± chlorite. Magnetite occurs as inclusions in garnetrystals which are commonly 1 mm in size and are euhedral tonhedral (Fig. 3c). Magnetite is also the most abundant oxide min-


ral associated with the magnetite-dominant stage and occurs asassive (Fig. 3d and e), disseminated (Fig. 3f), and brecciated ores

n the skarn at Sangan. It is found together with minor hematite,

gopite in magnesian skarn; (f) Magnetite occurs as disseminated in calcite grains.yroxene.

pyrite, and calcite (Fig. 3d) in the calcic skarn and with phlogo-pite (Fig. 3e) in the magnesian skarn. The massive ore shows fine-to medium-grained (up to 2.5 cm in diameter), granoblastic andmassive textures (Fig. 3e).

5. Results

5.1. Major element compositions of mineral

Chemical compositions and formulae of garnet grains from San-gan skarn types are reported in Table 1. Garnet compositions arepredominantly andradite-grossularite (And-Gro) (grandite) solidsolution, rich in Fe and Al, with spessartine component (Spe) lessthan 5 mol% (Fig. 4). Garnet from A skarn displays wide varia-tions in contents of andradite (And; 50.6–76.3%), grossular (Gro;22.1–47.3%), pyrope (Pyr; up to 0.4%), almandine (Alm; up to 0.9%)and spessartine (Spe; 1.3–1.9%, Table 1). Garnet from A’ has less Andcontent (40.8–55.4%), and more Gro content (39.7–55.2%) than thatof garnet from A skarn. Garnet from D skarn also shows wide vari-ation in And (57.3–74.0%), Gro (24.7–40.5%), Pyr (up to 0.5%) andAlm (up to 0.8%) contents, as does garnet from Cn (And; 48.9–68.3%,Gro; 29.8–48.1%, Pyr; up to 0.4%, Alm; up to 1.6% and Spe; 1.4–2.5%).Garnet from Cs skarn shows narrow variations in contents of And(50.9–60.9%), Gro (36.9–46.3%), Pyr (up to 0.15%), Alm (0.21–1.13%)and Spe (1.43–1.60%).

Garnets with Al-rich cores and Fe-rich rims are common in skarnsystem; however homogeneous garnets and garnets with Fe-richcores and Al-rich rims may also exist (Meinert, 1992, 1997). Withinthe Sangan skarns, some garnet crystals show subtle Fe+3 enrich-ment in rims relative to cores (Fig. 5). In general, garnetite samplesfrom different orebodies are intermediate grandites (Fig. 4) andanisotropic (Fig. 3a and b). Garnets of the A, A’ and Cn orebodiesformed by epitaxial growth with zoning and have trapezohedral


faces (Fig. 5a, d and g). These garnet grains show Fe-enriched andAl-depleted rims rather than cores (Fig. 5b, c, e, f h, i). Garnet grainsfrom Cs and D also show zoning (Fig. 5d, m) with intense fractures(Fig. 5j), but are not Fe-enriched and Al-depleted from rim to core


Please cite this article in press as: Sepidbar, F., et al., Mineral geochemistry of the Sangan skarn deposit, NE Iran: Implication for theevolution of hydrothermal fluid. Chemie Erde - Geochemistry (2017), http://dx.doi.org/10.1016/j.chemer.2017.07.008



Table 1(a) Selected electron microprobe analyses of garnet from A orebody. (b) Selected electron microprobe analyses of garnet from A’ orebody. (c) Selected electron microprobeanalyses of garnet from D orebody. (d) Selected electron microprobe analyses of garnet from Cn orebody. (e) Selected electron microprobe analyses of garnet from Cs orebody.

(a)

Sample Grt-A-011 Grt-A-04 Grt-A -05 Grt-A -06 Grt-A -07 Grt-A -08 Grt-A -09 Grt-A -010core core core core core rim rim rim

SiO2 37.98 37.74 37.44 38.02 37.96 37.79 37.03 37.24TiO2 0.348 0.251 0.163 0.29 0.41 0.051 0.109 0.017Al2O3 10.78 9.320 8.200 10.35 10.79 5.750 4.320 6.430FeO 15.64 17.43 18.62 16.26 15.95 22.24 23.35 21.55MnO 0.860 0.760 0.680 0.750 0.900 0.720 0.574 0.580MgO 0.080 0.100 0.060 0.072 0.042 0.000 0.091 0.030CaO 35.14 35.23 35.29 34.73 34.95 35.23 35.25 35.01

Number of cations on the basis of 12 oxygensSi 2.967 2.965 2.966 2.982 2.963 2.982 2.972 2.962Ti 0.020 0.015 0.010 0.017 0.024 0.003 0.006 0.001Al 0.992 0.862 0.765 0.957 0.992 0.534 0.408 0.602Cr 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000Fe3+ 1.016 1.145 1.234 1.044 1.016 1.468 1.567 1.422Fe2+ 0.005 0.000 0.000 0.023 0.025 0.000 0.000 0.011Mn 0.057 0.051 0.046 0.050 0.059 0.048 0.039 0.039Mg 0.009 0.012 0.007 0.008 0.005 0.000 0.011 0.003Ca 2.941 2.966 2.996 2.919 2.923 2.979 3.031 2.983Ura 0.000 0.000 0.00 0.000 0.000 0.000 0.000 0.000Pyr 0.310 0.390 0.23 0.280 0.160 0.000 0.350 0.120Spe 1.880 1.670 1.50 1.660 1.980 1.580 1.270 1.290Gro 47.03 41.22 37.57 45.11 46.44 25.68 22.10 27.98And 50.60 56.72 60.69 52.17 50.58 72.74 76.28 70.24Alm 0.180 0.000 0.000 0.780 0.840 0.000 0.000 0.370

(b)

Sample Grt-A’-011 Grt-A’-012 Grt-A’-013 Grt-A’-014 Grt-A’-015 Grt-A’-016 Grt-A’-017 Grt-A’-018rim rim rim core core core core core

SiO2 38.07 38.07 38.07 38.07 38.07 38.07 38.07 38.07TiO2 0.127 0.024 0.864 0.231 0.034 0.017 0.069 0.055Al2O3 9.695 8.995 12.94 10.80 10.38 9.985 10.61 9.878FeO 17.07 17.65 13.14 15.79 16.48 16.80 15.93 16.96MnO 0.874 0.913 1.170 0.898 0.920 0.968 0.975 0.973MgO 0.048 0.045 0.043 0.053 0.016 0.037 0.058 0.033CaO 34.98 34.76 34.66 35.46 35.06 35.22 34.96 34.91

Number of cations on the basis of 12 oxygensSi 2.981 2.981 2.973 2.970 2.988 2.977 2.990 2.992Ti 0.007 0.001 0.050 0.013 0.002 0.001 0.004 0.003Al 0.896 0.841 1.174 0.986 0.953 0.920 0.976 0.909Cr 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000Fe3+ 1.112 1.170 0.809 1.020 1.054 1.095 1.028 1.094Fe2+ 0.008 0.000 0.036 0.003 0.020 0.004 0.011 0.014Mn 0.058 0.061 0.076 0.059 0.061 0.064 0.064 0.064Mg 0.005 0.005 0.005 0.006 0.002 0.004 0.000 0.004Ca 2.940 2.952 2.859 2.941 2.927 2.950 2.923 2.922Ura 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000Gro 42.25 39.66 55.24 46.90 44.74 43.28 45.93 42.66And 55.37 58.14 40.82 50.85 52.51 54.34 51.32 54.60Spe 1.930 2.030 2.560 1.960 2.020 2.120 2.140 2.140Alm 0.270 0.000 1.210 0.090 0.670 0.120 0.380 0.470Pyr 0.190 0.18 0.170 0.200 0.060 0.140 0.220 0.130

(c)

Sample Grt-D-024 Grt-D-025 Grt-D-026 Grt-D-027 Grt-D-028 Grt-D-029 Grt-D-030rim rim rim rim rim Core Core

SiO2 37.36 37.36 37.36 37.36 37.36 37.36 37.36TiO2 0.031 0.131 0.000 0.019 0.000 1.678 1.712Al2O3 9.289 8.604 5.363 7.511 7.265 7.798 8.513FeO 17.84 18.56 22.37 20.05 20.08 18.18 17.58MnO 0.889 0.921 0.494 0.707 0.731 0.687 0.755MgO 0.000 0.055 0.039 0.066 0.061 0.112 0.093CaO 34.93 34.42 34.94 34.61 34.72 34.10 35.09

Number of cations on the basis of 12 oxygensSi 2.975 2.977 2.977 2.971 2.982 2.952 2.923Ti 0.002 0.008 0.000 0.001 0.000 0.101 0.101Al 0.861 0.803 0.506 0.705 0.681 0.733 0.791Cr 0.000 0.000 0.000 0.000 0.000 0.000 0.000





Table 1 (Continued)

(c)

Sample Grt-D-024 Grt-D-025 Grt-D-026 Grt-D-027 Grt-D-028 Grt-D-029 Grt-D-030rim rim rim rim rim Core Core

Fe3+ 1.154 1.206 1.499 1.314 1.331 1.213 1.159Fe2+ 0.019 0.023 0.000 0.021 0.004 0.000 0.000Mn 0.059 0.062 0.033 0.048 0.049 0.046 0.050Mg 0.000 0.006 0.005 0.008 0.007 0.013 0.011Ca 2.944 2.922 2.999 2.952 2.957 2.915 2.965Ura 0.000 0.000 0.000 0.000 0.000 0.000 0.000And 57.26 60.02 74.04 65.1 66.17 61.18 57.46Gro 40.13 36.94 24.71 32.38 31.83 36.81 40.51Alm 0.650 0.780 0.000 0.680 0.130 0.000 0.000Pyr 0.000 0.220 0.150 0.260 0.240 0.450 0.360Spe 1.960 2.050 1.100 1.570 1.630 1.560 1.670

(d)

Sample Grt-Cn-012 Grt-Cn-013 Grt-Cn-014 Grt-Cn-015 Grt-Cn-016 Grt-Cn-017 Grt-Cn-019 Grt-Cn-20 Grt-Cn-21rim rim rim rim rim core core core core

SiO2 38.02 38.02 38.02 38.02 38.02 38.02 38.02 38.02 38.02TiO2 0.010 0.002 0.041 0.170 0.017 0.444 0.535 0.356 0.285Al2O3 6.410 6.350 7.304 9.278 6.689 11.18 10.33 9.811 9.805FeO 20.93 21.02 20.74 18.15 20.63 15.73 16.79 17.00 17.69MnO 0.620 0.630 0.700 0.820 0.670 1.040 1.240 1.130 1.060MgO 0.038 0.025 0.037 0.055 0.064 0.100 0.076 0.081 0.099CaO 35.34 35.27 34.84 35.20 34.91 34.71 34.16 34.35 34.50

Number of cations on the basis of 12 oxygensSi 3.0074 2.980 2.981 2.962 2.980 2.963 2.962 2.992 2.954Ti 0.0007 0.000 0.002 0.010 0.001 0.026 0.031 0.021 0.017Al 0.5945 0.595 0.677 0.854 0.628 1.024 0.954 0.904 0.907Cr 0.0000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000Fe3+ 1.3774 1.398 1.334 1.164 1.375 0.983 1.051 1.087 1.113Fe2+ 0.0000 0.000 0.030 0.021 0.000 0.039 0.049 0.025 0.048Mn 0.0415 0.042 0.047 0.054 0.045 0.068 0.082 0.075 0.071Mg 0.0045 0.003 0.004 0.006 0.007 0.012 0.009 0.009 0.012Ca 2.9800 3.005 2.935 2.946 2.980 2.891 2.866 2.878 2.899Ura 0.00 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000And 68.28 68.73 66.34 57.68 68.00 48.98 52.42 54.59 55.11Gro 30.20 29.78 30.97 39.63 30.27 47.06 42.92 41.76 40.60Alm 0.00 0.000 1.000 0.700 0.000 1.300 1.640 0.830 1.570Pyr 0.15 0.100 0.140 0.210 0.250 0.380 0.300 0.320 0.380Spe 1.37 1.380 1.550 1.780 1.48 2.270 2.730 2.510 2.330

(e)

Grt-Cs-030 Grt-Cs-031 Grt-Cs-032Sample rim core rim

SiO2 37.54 38.36 37.84TiO2 0.000 0.070 0.149Al2O3 9.290 10.75 8.441FeO 18.32 16.27 18.85MnO 0.711 0.731 0.643MgO 0.000 0.014 0.006CaO 35.02 35.17 34.79

Number of cations on the basis of 12 oxygensSi 2.957 2.982 2.984Ti 0.000 0.004 0.009Al 0.862 0.985 0.785Cr 0.000 0.000 0.000Fe3+ 1.166 1.024 1.220Fe2+ 0.040 0.034 0.023Mn 0.047 0.048 0.043Mg 0.000 0.002 0.001Ca 2.955 2.930 2.940Ura 0.000 0.000 0.000And 57.5 50.97 60.86Gro 39.62 46.25 36.90Alm 1.330 1.130 0.780Pyr 0.000 0.050 0.020Spe 1.560 1.600 1.430




Fig. 4. Ternary diagram summarizing the compositional variations in garnet from different Sangan skarn zones. Abbreviations: Gro = grossular; And = andradite;Spe = almandine + spessartine. Diagram modified after Meinert et al. (2005). Garnet rim from A: filled violet circle; Garnet core from A: violet circle; Garnet rim fromA circlec blue ci

(S1l

stracarAMcfith(

lbaAbACM(dS

5

mweetpRhm

’: filled red circle; Garnet core from A’: red circle; Garnet rim from Cn: filled greenore from Cs: circle; Garnet rim from D: filled blue circle; Garnet core from D: lights referred to the web version of this article.)

Fig. 5k, l, n, o). Similar to other reported oscillatory grandites (e.g.,an Benito, Armbruster et al., 1998; Galore Creek, Russell et al.,999), garnet from the Sangan deposits are principally classified as

ow-Ti garnet, with Ti content less than 1 wt.%.Chemical compositions of clinopyroxene grains from Sangan

karns are reported in Table 2. In some places, pyroxene fromhe Sangan skarns shows zonation patterns with a rim more Fe-ich than the core (Mazaheri, 1996). Clinopyroxene from skarnssociated with A’ orebody shows a binary diopside-hedenbergiteomposition, ranging from Hd29 to Hd77 (Table 4). It has MgOnd FeO contents of 3.20–11.5 wt.% and 8.15–19.9 wt.% (Table 4),espectively. Other constituents, such as MnO (up to 0.85 wt.%),l2O3, and Na2O are generally less than 1.1 wt.% (Table 2) and then/Fe ratios are relatively restricted (<0.05). The composition of

linopyroxene crystals from the Sangan iron deposit plots in theeld of skarn deposits on the johanssonite-diopside-hedenbergite

ernary diagram (Fig. 6). Like with garnet, the clinopyroxeneas high Fe/Fe + Mg ratios (0.28–0.77) and low TiO2 (0.08), MnO0.19–0.89%) contents (Table 2).

Representative magnetite analyses from A, B, Cs and Cn areisted in Table 3 and presented in Fig. 7. Magnetite in the Cs ore-ody has high MgO (3.15–3.33 wt.%) and low SiO2 (<0.024 wt.%)nd MnO (0.09–0.13) contents. In addition, this magnetite has lowl2O3 (0.32–0.39 wt.%) content. Magnetite in the A, B, and Cn ore-odies contains moderate to low MgO (0.03–1.01 wt.% in sample-3; up to 0.03 wt.% in sample B-2, and 0.02–0.22 wt.% in samplen-1,) moderate to high SiO2 (0.06–3.43 wt.% for A, B, and Cn), lownO (0.01–0.12 wt.%, for A, B, and Cn) and low to moderate Al2O3

0.05–0.43 wt.%) contents (Table 3). In a Ca + Al + Mn versus Ti + Viscriminant diagram (Dupuis and Beaudoin 2011), magnetite fromangan plots dominantly in the skarn field (Fig. 7).

.2. Trace and rare earth element compositions

Garnet, clinopyroxene and magnetite from calc-silicate andagnetite dominated stages in the Sangan skarn mineralizationere analyzed for their trace element contents. Each of these min-

rals shows a range of characteristic patterns for REE. To somextent, these patterns reflect typical patterns for each mineral, withrends and LREE/HREE fractionation controlled by crystal structural


arameters intrinsic to each mineral group. Chondrite-normalizedEE patterns are summarized in Fig. 8. Garnet is the major REE-ost in the silicate dominated stage but a calculation for typicalulti-component skarn (garnet and pyroxene, 25%), using average

; Garnet core from Cn: green circle; Garnet rim from Cs: filled black circle; Garnetircle. (For interpretation of the references to color in this figure legend, the reader

REE concentrations in these minerals, shows that garnet accountfor >90% of �REE. Garnet grains from all of the skarn are char-acterized by similar REE patterns from core to rim; a depletion ofLREE relative to HREE and positive Eu anomalies (Fig. 8). The excep-tion is garnet from the Cn orebody where cores are more enrichedin LREE and depleted in HREE than the rim (Chondrite value Sunand McDonough, 1989). The results of LA-ICP-MS analyses showthat all garnet crystals in Sangan skarn are depleted in large ionlithophile elements (LILE; e.g., Rb, Ba, La, Sr) and enriched in highfield strength elements (HFSE; e.g., U, Th, Pb, Sm, Nd; Fig. 8; Table 4).Variations in LaN/YbN (chondrite normalized La/Nb ratios) versusFe3+/(Fe3+ + Al) have been discussed in the literature (e.g., Peng et al.,2014) to study of LREE/HREE fractionation between a hydrother-mal fluid and a growing garnet. Within the garnets from differentorebodies in Sangan (Fig. 9a), LREE/HREE fractionation does notincrease with increasing Fe contents from core to rim. Also, Eu/Eu*vs Fe3+/(Fe3+ + Al) ratios and �REE vs Al concentrations do not showany correlations (Fig. 9b, c). Al-rich garnet cores from A, Cs and Cnorebodies are generally more Zr-, Hf- and Nb-enriched, and Rb-,Ba- and Y-depleted than Fe-rich garnet rims, while garnets from Dorebodies generally show similar contents of these elements fromcore to rim. These results also show that garnet rims from A’ ore-body are Hf- enriched and Y-depleted compared to garnet cores(Table 4; Fig. 9d–i).

In-situ LA-ICP-MS analysis showed that clinopyroxene coreshave higher contents of REE compared to their rims (Fig. 10).They also have distinct REE patterns from core to rim. In general,clinopyroxene cores are HREE-rich, have relatively high �REE, low�LREE/�HREE ratios, as well as positive Eu anomalies (Fig. 10). Incontrast, pyroxene rims are LREE-rich, have relatively low �REE,high �LREE/�HREE ratios and lack Eu anomalies (Fig. 10; Table 5).

Magnetite grains from the Sangan skarns were analyzed forREE and trace elements by LA-ICP-MS. The results show that mag-netite from the Sangan Fe deposit contains substantial U, Ti, andCu, but these concentrations vary significantly, as illustrated bythe concentration of U (0.01–1.61 ppm), Ti (2.96–180.6 ppm), andCu (0.03–63.87 ppm), etc. In addition, there is a spatial variationwithin individual magnetite grains from Sangan, which has beenobserved in many iron deposits, such as the Los Colorados ironoxide-apatite deposit, Chile (Knipping et al., 2015), porphyry Cu-Mo-Au deposits in British Columbia, Canada (Canil et al., 2016) and


the Chengchao iron deposit, China (Hu et al., 2014) and others. Themagnetite grains from Sangan contain 0.03 to 64 ppm Cu, 1328 to5026 ppm Si, and ∼3 to 180 ppm Ti. The U values (∼0.01–1.6 ppm) of





Fig. 5. Garnet morphology and associated compositional variations from Sangan. Garnets from A and Cn orebodies show more Fe-enriched cores and Al-depleted rims thanthose from other types of skarn. Symbols as in Fig. 4.

Fig. 6. Johannsenite (Jo) – Diopside (Di) – Hedernbergite (Hd) ternary diagram (Meinert et al., 2005) showing the composition of pyroxenes from the Sangan A’ skarn irondeposit.




Table 2Selected electron micro probe analyses of pyroxene from Sangan orebody.

CPx-A’-1 CPx-A’-2 CPx-A’-3 CPx-A’-5 CPx-A’-6 CPx-A’-7 CPx-A’-8

SiO2 50.2 49.6 48.5 49.5 49.7 50.9 50.8TiO2 0.08 0.08 0.08 0.08 0.08 0.08 0.08Al2O3 0.10 0.09 0.18 0.10 0.60 1.30 0.55Cr2O3 0.10 0.18 0.11 0.11 0.16 0.10 0.10Fe2O3 4.30 2.60 3.36 4.90 4.99 3.38 3.49FeO 16.2 16.9 19.9 15.8 10.8 8.15 10.3MnO 0.19 0.69 0.85 0.34 0.42 0.28 0.32MgO 5.66 5.68 3.20 5.20 8.27 11.5 9.78CaO 22.4 23.7 23.2 22.5 24.8 23.2 24.5Na2O 1.06 0.25 0.40 1.05 0.35 0.44 0.30K2O 0.05 0.05 0.05 0.05 0.05 0.05 0.05Total 100 99.8 99.8 99.5 100 99.3 100Mn/Fe2+ 0.01 0.03 0.04 0.01 0.02 0.03 0.03CaMgSi2O6 33.2 33.7 19.3 31.0 48.2 64.2 55.8CaFeSi2O6 66.1 63.6 76.9 67.9 47.5 29.1 40.6CaMnSi2O6 0.64 2.30 2.90 0.00 1.39 0.89 1.04

a + M

m(gpsgc

5

oosmat(rstfl

Fig. 7. Position of magnetite samples from the Sangan on Al + C

agnetite grains (Table 6) are higher than to those from pyroxenesup to 0.08 ppm; Table 5) and significantly lower than those fromarnets (1.1–8.8 ppm) (Table 4; Fig. 13). Chondrite-normalized REEatterns for magnetite from the Sangan skarn are characterized bylight enrichment in LREEs relative to HREEs (Fig. 11a). Magnetiterains from the Cs orbody have Ba, Th, La, Nd, Sm, Y, Dy, Yb, and Luontents higher than those from other types of iron ores (Fig. 11b).

.3. Baro-acoustic decrepitation results of fluid inclusions

Fluid inclusions are trapped in garnet, quartz, calcite andpaque minerals of the Sangan skarn. To find the temperaturef hydrothermal fluid during garnet and quartz formation in theilicate-dominant stage we used the baro-acoustic decrepitationethod. One garnetite sample (A-1) was chosen from the skarn

ssociated with orebody A. The garnet has only low decrepi-ation intensity with two distinct peaks near 450◦ and 500 ◦CFig. 12) which are consistent with the results of microthermomet-


ic studies by Mazaheri (1996), from garnet (450–500 ◦C) in theilicate-dominant stage. It is not clear why there are 2 peaks onhe decrepigram, which may be due to different fluid events or touid inclusion size variations. Mazaheri (1996) described that fluid

n vs. Ti + V discriminant diagram (Dupuis and Beaudoin, 2011).

inclusions in garnet have wide variations in size and shape fromless than 3 �m to 10 �m and from rod –like, elongated to irregu-lar, respectively. The garnet formation temperature is estimated tobe about 450 ◦C and above. The baro-acoustic decrepitation datafrom the opaque minerals provides the best estimate of the forma-tion temperatures of these deposits (Burlinson, 1984). In this study,two magnetite samples (B-4 and C-4) were chosen from high-gradeorebodies B and C. The analysis of sample B-4 shows broad, lowintensity decrepitation with an onset temperature of 400 ◦C andpeaks at 450 ◦C and 600 ◦C. The decrepitation intensity of this sam-ple is low in comparison with magnetite from other skarn deposits(Burlinson, 2001), suggesting either a low abundance of fluid inclu-sions or later removal events such as recrystallization. The presenceof two separate peaks may indicate multiple fluid events but mayalso be caused by the size distribution of inclusions formed in asingle fluid event. The sample formation temperature is estimatedto be about 400 ◦C. The analysis of sample C-4 shows an intense,single, high temperature peak at 790 ◦C. The formation tempera-

◦


ture of sample C-4 is also estimated to be about 400 C. There wereinsufficient samples in this study to understand the significance ofthe different shapes of the decrepigram peaks or the dual peaks insamples A-1 and B-4. However it is notable that no samples had





Table 3(a) Selected electron micro probe analyses of magnetite from Cn orebody. (b) Selected electron micro probe analyses of magnetite from Cs orebody. (c) Selected electronmicro probe analyses of magnetite from A orebody. (d) Selected electron micro probe analyses of magnetite from B orebody.

(a)

Sample Mt-Cn Mt-Cn Mt-Cn Mt-Cn Mt-Cn Mt-Cn Mt-Cn Mt-Cn Mt-Cn Mt-CnNumber Cn-1-1 Cn-1-2 Cn-1-3 Cn-1-4 Cn-1-5 Cn-1-6 Cn-1-7 Cn-1-8 Cn-1-9 Cn-1-10

SiO2 0.635 0.814 0.309 0.19 0.692 0.105 0.123 0.142 0.135 0.062TiO2 0.052 0.053 0.053 0.035 0.051 0.110 0.137 0.098 0.054 0.047Al2O3 0.258 0.283 0.315 0.191 0.295 0.286 0.276 0.253 0.162 0.118Cr2O3 0.001 0.002 0.001 0.003 0.001 0.001 0.006 0.007 0.002 0.002V2O3 0.010 0.011 0.002 0.001 0.002 0.010 0.017 0.012 0.023 0.005Fe2O3 67.53 66.58 67.52 67.74 67.78 69.40 68.77 67.71 68.82 68.64FeO 31.71 31.53 31.00 30.94 31.91 31.78 31.50 31.11 31.42 31.09MnO 0.025 0.046 0.051 0.027 0.034 0.060 0.039 0.015 0.017 0.028MgO 0.227 0.261 0.103 0.074 0.225 0.033 0.072 0.019 0.035 0.043CaO 0.032 0.109 0.080 0.020 0.071 0.031 0.002 0.004 0.001 0.003ZnO 0.000 0.018 0.000 0.000 0.016 0.000 0.013 0.000 0.002 0.006Total 100.4 99.72 99.53 99.23 101.0 101.8 100.9 99.37 100.7 100.1

Number of cations on the basis of 32 oxygensSi 0.194 0.249 0.095 0.059 0.209 0.031 0.037 0.043 0.041 0.019Ti 0.012 0.052 0.052 0.011 0.011 0.024 0.031 0.022 0.012 0.011Al 0.127 0.102 0.114 0.069 0.10 0.101 0.190 0.122 0.078 0.063Cr 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.001 0.000 0.000V 0.002 0.002 0.000 0.001 0.000 0.002 0.004 0.003 0.006 0.001Fe(iii) 15.49 15.37 15.66 15.79 15.45 15.78 15.75 15.77 15.83 15.89Fe(ii) 8.085 8.090 8.020 8.0191 8.084 8.016 8.022 8.053 8.032 8.000Mn 0.006 0.011 0.013 0.010 0.017 0.015 0.010 0.004 0.014 0.013Zn 0.000 0.00 0.000 0.000 0.003 0.000 0.002 0.000 0.000 0.001Mg 0.103 0.119 0.047 0.034 0.101 0.014 0.032 0.009 0.016 0.019Ca 0.030 0.045 0.046 0.026 0.030 0.040 0.035 0.023 0.023 0.001Si 0.194 0.249 0.095 0.058 0.209 0.031 0.037 0.044 0.041 0.019Al 0.127 0.10 0.114 0.069 0.105 0.101 0.190 0.78 0.06 0.123Ti/V 6.000 26.00 – 11.00 – 12.00 8.000 22.00 2.000 –Mt-Cn: Magnetite from Cn orebody;

(b)

Sample Mt-Cs Mt-Cs Mt-Cs Mt-Cs Mt-Cs Mt-Cs Mt-Cs Mt-Cs Mt-Cs Mt-CsNumber Cs-3-1 Cs-3-2 CS-3-3 CS-3-4 CS-3-5 CS-3-6 CS-3-7 CS-3-8 CS-3-9 CS-3-10

SiO2 0.024 0.009 0.014 0.005 0.000 0.018 0.012 0.022 0.003 0.003TiO2 0.010 0.011 0.013 0.014 0.010 0.025 0.007 0.008 0.011 0.011Al2O3 0.393 0.358 0.360 0.371 0.322 0.343 0.343 0.327 0.322 0.322Cr2O3 0.000 0.000 0000 0.000 0.000 0.000 0.004 0.003 0.005 0.005V2O3 0.000 0.018 0.012 0.007 0.008 0.008 0.010 0.016 0.002 0.002Fe2O3 70.74 71.16 71.07 71.03 70.96 71.67 71.28 70.58 70.05 70.05FeO 26.39 26.22 26.29 26.16 26.29 26.62 26.44 26.10 25.73 25.73MnO 0.121 0.128 0.112 0.124 0.132 0.141 0.144 0.126 0.092 0.092MgO 3.153 3.302 3.274 3.333 3.201 3.234 3.207 3.231 3.327 3.327CaO 0.025 0.051 0.032 0.025 0.032 0.035 0.041 0.052 0.010 0.010ZnO 0.011 0.019 0.007 0.002 0.008 0.000 0.005 0.004 0.000 0.001Total 100.8 101.2 101.1 101.0 100.9 102.1 101.5 100.4 99.56 99.55

Number of cations on the basis of 32 oxygensSi 0.007 0.002 0.004 0.001 0.000 0.005 0.003 0.006 0.000 0.000Ti 0.002 0.002 0.002 0.003 0.002 0.005 0.001 0.001 0.002 0.002Al 0.137 0.124 0.125 0.129 0.112 0.118 0.119 0.115 0.114 0.114Cr 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.001 0.001V 0.000 0.004 0.002 0.001 0.001 0.001 0.002 0.003 0.000 0.000Fe(iii) 15.84 15.86 15.85 15.85 15.88 15.85 15.86 15.86 15.87 15.87Fe(ii) 6.569 6.494 6.519 6.490 6.537 6.547 6.540 6.520 6.482 6.482Mn 0.050 0.032 0.028 0.031 0.033 0.035 0.036 0.031 0.023 0.023Zn 0.002 0.004 0.001 0.000 0.001 0.000 0.001 0.001 0.000 0.000Mg 1.399 1.457 1.447 1.474 1.419 1.417 1.414 1.438 1.493 1.493Ca 0.008 0.016 0.010 0.007 0.010 0.011 0.013 0.016 0.003 0.003Si 0.007 0.002 0.004 0.001 0.000 0.005 0.003 0.006 0.000 0.000Al 0.137 0.124 0.125 0.129 0.112 0.118 0.119 0.115 0.114 0.114Ti/V – 0.500 1.000 3.000 2.000 5.000 0.500 0.250 – –Mt-Cs: Magnetite from Cs orebody

(c)

Sample Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-ANumber A-3-1 A-3-2 A-3-3 A-3-4 A-3-5 A-3-6 A-3-7 A-3-8 A-3-9 A-3-10 A-3-11 A-3-12

SiO2 0.314 0.520 0.193 1.148 0.474 1.399 3.432 0.314 0.520 0.193 0.835 1.127TiO2 0.032 0.027 0.034 0.021 0.055 0.041 0.000 0.032 0.027 0.034 0.000 0.000Al2O3 0.056 0.054 0.050 0.074 0.055 0.116 0.153 0.056 0.054 0.050 0.096 0.080Cr2O3 0.000 0.011 0.000 0.000 0.006 0.004 0.001 0.000 0.011 0.000 0.006 0.000




Table 3 (Continued)

(c)

Sample Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-A Mt-ANumber A-3-1 A-3-2 A-3-3 A-3-4 A-3-5 A-3-6 A-3-7 A-3-8 A-3-9 A-3-10 A-3-11 A-3-12

V2O3 0.004 0.015 0.007 0.007 0.005 0.002 0.005 0.004 0.015 0.007 0.000 0.002Fe2O3 67.59 68.09 68.79 65.71 68.05 65.71 61.49 67.59 68.09 68.79 67.67 66.19FeO 30.85 31.40 30.98 31.76 31.44 32.19 33.61 30.85 31.40 30.98 31.26 31.24MnO 0.091 0.090 0.071 0.098 0.081 0.085 0.120 0.091 0.090 0.071 0.080 0.102MgO 0.036 0.056 0.071 0.160 0.061 0.264 1.013 0.036 0.056 0.071 0.134 0.160CaO 0.200 0.298 0.261 0.200 0.199 0.190 0.348 0.200 0.298 0.261 0.728 0.699ZnO 0.000 0.000 0.003 0.000 0.008 0.004 0.000 0.000 0.000 0.003 0.000 0.001Total 99.17 100.5 100.4 99.19 100.4 99.97 100.1 99.17 100.5 100.4 100.8 99.61

Number of cations on the basis of 32 oxygensSi 0.097 0.158 0.059 0.354 0.144 0.427 1.029 0.097 0.158 0.059 0.253 0.345Ti 0.008 0.006 0.007 0.004 0.012 0.008 0.000 0.007 0.006 0.007 0.000 0.000Al 0.020 0.019 0.018 0.026 0.019 0.041 0.054 0.020 0.019 0.018 0.034 0.028Cr 0.000 0.002 0.000 0.000 0.001 0.000 0.000 0.000 0.002 0.000 0.001 0.000V 0.001 0.003 0.001 0.001 0.001 0.002 0.001 0.000 0.003 0.001 0.000 0.000Fe(iii) 15.76 15.64 15.84 15.2 15.66 15.10 13.88 15.76 15.64 15.84 15.45 15.27Fe(ii) 7.997 8.018 7.929 8.193 8.041 8.222 8.434 7.997 8.018 7.929 7.935 8.016Mn 0.023 0.023 0.018 0.025 0.020 0.021 0.030 0.023 0.023 0.018 0.020 0.026Zn 0.000 0.000 0.000 0.000 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000Mg 0.016 0.025 0.032 0.073 0.027 0.120 0.453 0.016 0.025 0.032 0.060 0.073Ca 0.066 0.097 0.085 0.066 0.065 0.062 0.111 0.066 0.097 0.085 0.236 0.229Si 0.097 0.158 0.059 0.354 0.144 0.427 1.029 0.097 0.158 0.059 0.253 0.345Al 0.020 0.019 0.018 0.026 0.019 0.041 0.054 0.020 0.019 0.018 0.034 0.028Ti/V 8.000 2.000 7.000 4.000 11.00 4.000 0.000 7.000 2.000 7.000 0.000 0.000Mt-A: Magnetite from A orebody

(d)

Sample Mt-B Mt-B Mt-B Mt-B Mt-B Mt-B Mt-B Mt-B Mt-B Mt-B Mt-B Mt-BNumber B-2-1 B-2-2 B-2-3 B-2-4 B-2-5 B-2-6 B-2-7 B-2-8 B-2-9 B-2-10 B-2-11 B-2-12

SiO2 0.779 1.393 1.393 0.615 1.688 1.500 1.318 1.153 0.694 0.846 0.906 0.915TiO2 0.002 0.001 0.001 0.001 0.001 0.001 0.034 0.049 0.000 0.000 0.000 0.000Al2O3 0.392 0.439 0.439 0.221 0.377 0.341 0.161 0.164 0.137 0.248 0.248 0.225Cr2O3 0.000 0.000 0.000 0.000 0.001 0.007 0.000 0.001 0.000 0.004 0.004 0.000V2O3 0.001 0.001 0.001 0.000 0.003 0.002 0.000 0.000 0.000 0.000 0.000 0.009Fe2O3 66.29 64.92 64.92 67.22 64.83 64.83 65.77 65.76 68.47 66.99 67.02 68.03FeO 31.86 32.71 32.71 31.67 33.05 32.80 32.74 32.30 32.25 32.06 32.26 32.73MnO 0.021 0.038 0.038 0.028 0.041 0.026 0.030 0.038 0.027 0.050 0.029 0.042MgO 0.000 0.006 0.006 0.003 0.023 0.027 0.015 0.016 0.000 0.018 0.006 0.018CaO 0.065 0.065 0.065 0.125 0.251 0.098 0.095 0.125 0.215 0.154 0.149 0.115ZnO 0.000 0.000 0.000 0.003 0.016 0.000 0.000 0.000 0.001 0.000 0.000 0.000Total 99.42 99.57 99.57 99.90 100.2 99.63 100.1 99.58 101.8 100.3 100.6 102.0

Number of cations on the basis of 32 oxygensSi 0.240 0.427 0.427 0.188 0.513 0.459 0.402 0.354 0.209 0.258 0.275 0.275Ti 0.001 0.001 0.001 0.0001 0.002 0.001 0.007 0.011 0.008 0.003 0.002 0.001Al 0.142 0.158 0.158 0.080 0.135 0.123 0.057 0.059 0.049 0.089 0.089 0.079Cr 0.000 0.000 0.000 0.000 0.000 0.001 0.000 0.000 0.000 0.001 0.000 0.001V 0.001 0.001 0.001 0.001 0.001 0.001 0.002 0.001 0.001 0.001 0.001 0.002Fe(iii) 15.37 14.98 14.98 15.54 14.83 14.95 15.12 15.21 15.53 15.39 15.35 15.37Fe(ii) 8.213 8.393 8.393 8.138 8.407 8.408 8.364 8.307 8.133 8.187 8.217 8.219Mn 0.0052 0.009 0.009 0.007 0.010 0.006 0.007 0.019 0.019 0.013 0.014 0.011Zn 0.000 0.000 0.000 0.000 0.003 0.000 0.000 0.013 0.000 0.012 0.003 0.011Mg 0.0000 0.002 0.002 0.001 0.010 0.012 0.018 0.041 0.069 0.050 0.049 0.037Ca 0.021 0.021 0.021 0.041 0.081 0.032 0.031 0.355 0.209 0.258 0.276 0.275

0.4591.000

lbtv

6

maDf

Si 0.240 0.427 0.427 0.188 0.513

Ti/V 1.000 1.000 1.000 1.000 2.000

Mt-B: Magnetite from B orebody

ow temperature decrepitation (below 300 ◦C) commonly causedy CO2 rich fluids (Burlinson, 1984, 2012) and therefore we ten-atively suggest that these deposits were formed from fluids withery low CO2 contents (Fig. 12).

. Discussion

As discussed in Section 5.1, major element compositions of


inerals, the garnet and pyroxene from the Sangan skarn arendradite-grossularite, and diopside-hedenbergite, respectively.ue to their similar Mg/(Mg + Fe) ratios (1.00–1.08 and 1.00–1.50

or garnet and pyroxene, respectively), it is inferred that these

0.402 0.059 0.049 0.089 0.089 0.079 3.500 11.00 8.000 3.000 2.000 1.000

minerals were formed in equilibrium at the same tempera-tures. Using the garnet-clinopyroxene geothermometer based ongarnet-clinopyroxene Fe2+-Mg2+ distribution between garnet andpyroxene (Ravna, 2000), it is calculated that these mineral pairsin the Sangan skarn equilibrated at temperatures of 500–600 ◦C.The calc-silicate dominant stage was formed at temperaturesaround 350–400 ◦C as indicated by mineral chemistry and thedecrepitation analysis of garnet sample A-1 (Fig. 12). In addi-


tion, decrepitation data of the magnetite samples in the samediagram (Fig. 12) show that sample B-4 has broad, low intensitydecrepitation with peaks at 450 ◦C and 600 ◦C, perhaps indicat-ing a multi-stage fluid event. The decrepitation analysis of sample





Table 4Selected LA-ICP-MS analyses of garnet from Sangan skarn.

Sample (ppm) A-011 A-04 A-05 A-06 A-07 A-08 A-09 A-010core core core core core rim rim rim

Ca 230798.5 231921.7 228343.1 232377.2 224126.5 226408.2 227266.6 229988.9Sc 5.378138 5.467443 2.203846 1.327666 1.138141 1.298890 0.506061 1.312782Ti 5179.811 5230.487 3534.300 2344.550 1327.924 641.9432 128.6585 3566.747Fe3+ 100421.1 102235.1 97441.48 100496.6 94172.54 135487.3 140590.3 104332.9Fe2+ 97352.70 97669.33 93053.17 96677.15 90818.68 130725.9 135783.1 100330.8Cu 0.112127 0.112127 0.116096 0.106174 0.110143 0.090297 0.101212 0.117089Rb 0.248069 0.140407 0.146857 0.015281 0.030264 0.026097 0.091091 0.069459Sr 2.341772 0.391949 0.779929 0.054178 0.187540 0.281806 0.53186 0.829543Y 17.24576 17.14653 15.22152 16.05503 10.77612 42.78695 7.203926 16.80916Zr 101.4007 106.759 136.2197 99.13833 60.31055 38.54001 21.01641 97.49115Nb 18.62503 19.73638 17.73198 15.83673 11.26234 12.07600 7.104698 21.43317Mo 1.552912 1.295913 1.410025 1.379264 1.287975 1.274083 1.268129 1.307820Ba 4.901845 0.397903 1.49933 0.052591 0.129988 0.207386 0.188533 0.309590La 0.028379 0.048622 0.066284 0.090793 0.132767 0.092282 0.052491 0.458432Ce 0.310582 0.308598 0.630095 1.08555 1.364380 0.937701 0.459424 3.373739Pr 0.118180 0.141598 0.258984 0.425687 0.502092 0.425687 0.258984 1.075628Nd 1.590619 1.699769 2.679146 4.157638 4.614085 5.487288 4.108024 9.079328Sm 1.554897 1.668016 1.970661 2.679146 2.470768 5.85443 6.578792 3.711113Eu 1.429870 1.455669 2.168124 2.460845 2.579918 4.127869 5.358292 2.589841Gd 2.391386 2.460845 2.421154 2.808142 2.282236 7.442072 8.225971 3.314203Tb 0.485223 0.451486 0.401872 0.478277 0.364165 1.218515 0.929763 0.539798Dy 3.006597 3.01652 2.718837 2.937138 2.292158 7.610759 3.274512 3.026443Ho 0.656887 0.634065 0.577505 0.588420 0.400880 1.609472 0.319513 0.637041Er 1.891279 1.826781 1.554897 1.651148 1.082573 4.435475 0.468354 1.655117Tm 0.277837 0.27883 0.236162 0.2441 0.168389 0.603304 0.048026 0.226239Yb 1.964707 1.913109 1.634279 1.6958 1.099442 3.959183 0.303637 1.733507Lu 0.296691 0.270891 0.235169 0.238146 0.152414 0.480262 0.041576 0.218301Hf 5.854430 5.844508 5.139991 2.629532 1.35148 0.784891 0.335389 2.292158Ta 2.530305 2.560073 2.033174 0.829543 0.545752 0.385995 0.061918 2.450923W 3.750804 3.711113 6.469642 6.777247 8.58319 1.159971 0.251046 17.90066204Pb 0.922817 1.061736 0.972431 0.952585 1.012122 1.002199 0.873203 0.992276206Pb 0.248069 0.452478 0.511022 0.074421 0.096251 0.071444 0.029768 0.240131207Pb 0.173648 0.437594 0.433625 0.033737 0.029768 0.069459 0.029768 0.071444208Pb 0.171664 0.377065 0.470339 0.017563 0.022624 0.04753 0.022227 0.031654232Th 0.036417 0.03979 0.036714 0.098731 0.067971 0.441563 0.161543 0.112127238U 2.877601 2.778374 3.324126 3.522581 3.552349 1.30782 0.832520 8.821336

Sample (ppm) CN-013 CN-014 CN-015 CN-016 CN-017 CN-018 CN-019 CN-020rim rim rim rim core core core core

Ca 233474 232823.6 230633.4 228321.6 223670.2 225049.7 227103.7 230992.8Sc 1.736889 0.324166 0.851436 3.381731 2.645354 3.791941 3.081577 4.562335Ti 1793.688 48.85501 1176.642 3748.969 2541.631 2065.747 1593.736 1690.625Fe3+ 126210.1 128001.7 130970.3 131599 126137.5 125807.8 128834.9 134190Fe2+ 122688.2 124730.5 127524 128622.4 123645.7 123603 127591.8 132702.8Cu 0.09705 0.115059 0.11706 0.108055 0.119061 0.104053 0.101052 0.100051Rb 0.016508 0.017909 0.083143 0.031316 0.016709 0.014507 0.016609 0.015908Sr 0.070736 0.089246 0.124564 0.099551 0.065233 0.046724 0.055729 0.063132Y 41.37118 21.79115 20.10029 32.79679 15.45791 14.82759 10.90558 10.87557Zr 27.41403 1.278654 18.86966 30.35554 68.43503 88.15512 89.24568 119.191Nb 12.96664 5.092607 4.722417 10.46536 7.493836 5.903021 6.163155 7.093631Mo 1.151589 0.926474 1.314673 1.382708 1.568803 1.611825 1.309670 0.99651Ba 0.05803 0.115059 0.121062 0.075038 0.05903 0.046024 0.062032 0.047024La 0.005903 0.007404 0.008204 0.008604 0.010906 0.017109 0.041521 0.046524Ce 0.056629 0.043822 0.088545 0.096449 0.341175 0.619317 0.768393 0.965494Pr 0.047224 0.032016 0.050526 0.05823 0.174589 0.345177 0.427219 0.502257Nd 0.779399 0.505259 0.973498 1.012518 2.941506 5.582858 6.133139 7.533856Sm 2.191122 1.074550 2.191122 1.954000 3.171623 4.342223 4.882499 4.892504Eu 1.263647 0.274140 1.221625 1.483759 2.46126 2.451255 2.391224 2.241147Gd 6.293221 2.551306 4.522315 3.811951 3.741915 4.062079 3.781936 3.781936Tb 1.384709 0.647331 0.803411 0.757388 0.627321 0.620318 0.491251 0.494253Dy 8.344271 4.102100 4.402253 4.932525 3.271675 3.121598 2.541301 2.611337Ho 1.554796 0.824422 0.737377 1.185607 0.572293 0.54728 0.403206 0.428219Er 3.611849 1.943995 1.700871 3.271675 1.298665 1.241636 0.884453 1.000512Tm 0.459235 0.228117 0.214110 0.517265 0.165985 0.175090 0.123063 0.119361Yb 2.971521 1.466751 1.395714 3.69189 1.091559 1.182605 0.778398 0.803411Lu 0.426218 0.193099 0.179492 0.554284 0.150177 0.162083 0.105354 0.097750Hf 0.805412 0.021111 0.458235 1.869957 2.421239 2.941506 2.271162 3.341710Ta 0.357183 0.008004 0.187096 0.780399 0.693355 0.598306 0.339174 0.370189W 0.806413 1.185607 0.214110 0.195100 0.033017 0.021811 0.039920 0.036018204Pb 0.870446 0.980502 0.850435 0.920471 0.880451 1.080553 0.980502 0.790405206Pb 0.046024 0.034818 0.060031 0.062032 0.037019 0.050026 0.107055 0.108055207Pb 0.047524 0.049425 0.049025 0.039020 0.034017 0.027014 0.037019 0.088045208Pb 0.014007 0.015108 0.065734 0.061832 0.017909 0.015808 0.013107 0.018309232Th 0.065233 0.043722 0.029315 0.030516 0.159081 0.175090 0.186095 0.253130238U 0.565289 0.049025 0.238122 1.160594 1.432733 2.151101 2.321188 2.981526





Table 4 (Continued)

Sample (ppm) D-024 D-025 D-026 D-027 D-028 D-029 D-030

Ca 227545.3 229870.1 232009.6 230531.2 234830.6 231380.7 229319.2Sc 0.493615 0.388402 0.535897 0.389386 0.357920 0.506398 0.343171Ti 599.8802 395.9542 525.6904 382.5617 385.4231 378.0975 430.0353Fe3+ 136091.9 137319.3 138920.6 136681.7 136194.2 137118.2 1383740Fe2+ 135587.6 136708.7 138385.0 136762.3 136404.6 137176.0 138467.2Cu 0.099313 0.091447 0.095380 0.115046 0.101280 0.096363 0.099313Rb 0.061456 0.016323 0.015143 0.014946 0.026942 0.013668 0.082105Sr 0.329405 0.048575 0.079352 0.031269 0.073846 0.036677 0.318588Y 23.49097 21.88820 23.68763 21.59321 22.12419 20.84590 24.82826Zr 24.71026 15.02478 20.16743 13.90382 13.27451 14.37581 11.62257Nb 8.741514 7.659887 7.866379 7.276401 6.558593 3.638200 5.653960Mo 0.970515 0.984281 1.091460 1.062944 1.209456 1.090477 0.968548Ba 1.594908 0.058998 0.036382 0.040315 0.062931 0.058998 1.097360La 0.053688 0.055163 0.053000 0.052606 0.080040 0.039135 0.048280Ce 0.933149 1.128825 0.874151 1.003947 0.873168 0.748289 0.863335Pr 0.503448 0.637177 0.496565 0.553596 0.512298 0.407085 0.480832Nd 7.345231 8.830010 7.178071 7.974542 7.600889 6.332435 7.089574Sm 7.581223 8.426858 7.453394 8.338362 7.591056 6.371767 7.178071Eu 6.037446 6.430765 5.929283 6.293103 5.948949 5.358971 5.634294Gd 7.207570 7.335398 7.296066 7.266568 7.522225 6.155442 6.322602Tb 1.004930 0.965598 0.993130 1.000997 0.972481 0.859402 0.891851Dy 5.014817 4.877155 5.181977 4.857489 4.906654 4.424838 4.719827Ho 0.892834 0.824986 0.847602 0.828920 0.843669 0.751239 0.870218Er 1.946929 1.740436 1.907597 1.81910 1.730603 1.819100 2.212419Tm 0.240908 0.228125 0.225175 0.205509 0.226158 0.209442 0.267457Yb 1.390382 1.275337 1.304836 1.252721 1.272387 1.346134 1.622441Lu 0.167161 0.147691 0.176994 0.151428 0.164211 0.161261 0.229108Hf 0.061259 0.016618 0.129795 0.030974 0.019371 0.028024 0.018683Ta 0.067454 0.033727 0.060276 0.030581 0.033825 0.037169 0.039037W 0.533930 0.578179 0.291056 0.494599 0.570312 0.077680 0.374636204Pb 0.963631 0.737473 0.776805 0.894801 0.865302 0.894801 0.934133206Pb 0.214359 0.160277 0.141595 0.196659 0.402169 0.101280 0.922333207Pb 0.104230 0.031466 0.080630 0.084564 0.335304 0.026549 0.819087208Pb 0.097445 0.015929 0.049853 0.082499 0.281223 0.013570 0.796471232Th 0.109441 0.135695 0.099116 0.100296 0.072567 0.047395 0.103050238U 5.496632 6.352101 4.788658 5.516298 5.103313 3.402209 4.631331

Sample (ppm) A’-011 A’-012 A’-013 A’-014 A’-015 A’-016 A’-017 A’-018rim rim rim core core core core core

Ca 226252.0 227004.7 229023.2 228116.5 233707.4 232994.4 230432.8 233002.4Sc 0.973857 0.567082 0.978867 2.028869 2.253297 6.562516 12.38362 9.628363Ti 1038.721 4217.053 2204.965 115.3700 134.8372 125.2388 521.5948 279.0021Fe3+ 113275.9 101647.3 107986.4 120303.5 123494.5 127825.0 125579.8 126800.3Fe2+ 111783.5 100354.2 106603.2 118684.1 121833.9 125962.3 124113.8 124779.7Cu 0.090172 0.096183 0.104199 0.100191 0.082157 0.095182 0.113216 0.101193Rb 0.024547 0.058912 0.014127 0.017834 0.013125 0.012624 0.020038 0.016031Sr 0.079051 0.140869 0.065124 0.028154 0.042481 0.043082 0.127543 0.163512Y 17.50338 33.81449 19.00625 13.17513 14.77818 9.648401 13.06492 11.24144Zr 122.4034 114.7488 109.1782 100.0107 134.3462 108.9879 129.7875 162.2895Nb 13.26530 31.68042 18.52533 5.610700 7.965191 3.025771 8.085420 5.710892Mo 1.183257 1.405681 1.552962 1.446759 1.246377 1.228343 1.407685 1.224335Ba 0.048092 0.168321 0.055105 0.039075 0.047090 0.051097 0.151289 0.051097La 0.065826 0.047791 0.036570 0.033263 0.090372 0.027252 0.043683 0.059213Ce 0.672282 0.726385 0.493942 0.556060 1.067035 0.632206 0.932779 1.034974Pr 0.313598 0.342653 0.237453 0.318608 0.504963 0.327625 0.458875 0.530011Nd 4.067758 4.759076 3.276248 4.338274 6.572535 4.789134 6.121675 6.802974Sm 3.697051 4.338274 3.186076 4.047720 5.430356 4.187987 4.729019 5.420337Eu 2.995713 3.166038 2.552869 2.975675 3.707070 3.446573 3.446573 4.087796Gd 3.767185 4.989516 3.646955 3.867376 4.729019 4.177968 4.458503 4.809172Tb 0.609162 0.862645 0.564076 0.566080 0.652244 0.567082 0.590125 0.631204Dy 3.456592 5.470433 3.366420 2.955637 3.376439 2.694138 3.005732 2.985694Ho 0.667273 1.191272 0.673284 0.516986 0.578103 0.408780 0.493942 0.452864Er 1.689222 3.386458 1.936694 1.201291 1.300480 0.806538 1.251387 0.952817Tm 0.248875 0.509973 0.294562 0.166818 0.157300 0.091174 0.154394 0.112615Yb 1.654155 3.797242 1.984785 1.000909 1.024955 0.649238 0.999907 0.713360Lu 0.204089 0.509973 0.272520 0.138965 0.139967 0.093178 0.130048 0.091675Hf 2.604968 2.533832 2.358498 2.445664 2.905541 2.584930 3.957548 3.526726Ta 0.287548 1.239364 0.722378 0.021541 0.019136 0.005210 0.049795 0.023345W 3.045809 4.729019 3.396478 0.444848 2.755255 0.590125 2.735216 4.228064204Pb 0.821567 0.761452 0.801529 1.041987 0.871662 0.801529 0.971853 0.701338206Pb 0.106203 0.221422 0.111212 0.037071 0.094180 0.071236 0.115220 0.091174207Pb 0.040076 0.070134 0.033063 0.048392 0.032061 0.031059 0.042080 0.039976208Pb 0.017934 0.037872 0.021641 0.018836 0.020940 0.016131 0.022042 0.015530232Th 2.200196 0.094981 1.758353 0.571089 2.584930 0.886691 2.835408 2.060930238U 3.005732 6.111656 2.875484 1.309497 2.875484 1.499860 3.035790 3.316325




Table 4 (Continued)

Sample (ppm) Cs-030 Cs-031 Cs-032rim core rim

Ca 232828.4 226401.8 233225.5Sc 0.941367 13.11332 0.425809Ti 145.5927 420.0051 149.6713Fe3+ 132691.8 113340.9 132422.3Fe2+ 130230.9 111618.6 131121.3Cu 0.091743 0.089749 0.088752Rb 0.046769 0.013861 0.011767Sr 0.097128 0.040686 0.058636Y 30.69414 16.34428 22.24777Zr 20.76192 44.71492 43.95704Nb 1.770049 4.078591 4.597141Mo 0.98325 1.164742 1.041088Ba 0.139609 0.051855 0.050858La 0.005584 0.005285 0.011368Ce 0.047268 0.033506 0.059234Pr 0.033805 0.035800 0.040387Nd 0.587357 0.551457 0.880537Sm 2.009379 1.692266 2.473082Eu 1.694261 0.975272 2.038298Gd 6.272454 4.178312 6.800976Tb 1.248508 0.749902 1.234547Dy 7.249721 4.048675 6.212622Ho 1.160753 0.644198 0.905467Er 2.662552 1.415042 1.768054Tm 0.296172 0.170024 0.191365Yb 1.657364 1.060035 1.071004Lu 0.195653 0.127842 0.120064Hf 0.734944 1.572601 1.472880Ta 0.024132 0.071201 0.025728W 0.196450 0.609296 0.754888204Pb 0.797768 0.777824 0.907462206Pb 0.084763 0.066813 0.026127207Pb 0.035900 0.033905 0.036498208 .03350

.04447

.18548

Ciwtflsd(

ptacaiaaaf

onoh

6m

lc

Pb 0.038492 0232Th 0.007379 0238U 0.216395 0

-4 shows a single, intense, high temperature peak at 790 ◦C, thenterpretation of which is unclear. The magnetite dominant stage

as formed at a temperature of about 400 ◦C. Lack of any low-emperature peak around 300 ◦C, shows an absence of CO2-richuid inclusions (Sosnicka et al., 2015) which is consistent withtudies by Mazaheri (1996) who showed that garnet from a silicate-ominant stage was formed from hydrothermal fluid with low CO2XCO2 < 0.07) content.

Anisotropic garnets from Sangan have a wide range in com-osition (Fig. 4) mainly reflecting major-element heterogeneity ofhe protolith. Gaspar et al. (2008) showed that variations within

single sample are due to the existence of chemical potentialsontrolled by fluid pathways at small scale (centimeter or less). Inddition, minerals from skarn are able to record numerous geolog-cal processes including variations of physicochemical conditionsnd hydrothermal fluid evolution. In this contribution, major, tracend rare earth element (REE) concentrations of garnet, pyroxenend magnetite from the Sangan iron skarn deposit are importantor the investigation of hydrothermal fluid evolution.

In the following section, we consider the origin and evolutionf hydrothermal fluids responsible for garnet, pyroxene and mag-etite formation based on (1) Controls on compositional variationf garnets, pyroxene and magnetite, and (2) Implications for theydrothermal fluid evolution of the skarn.

.1. Controls on compositional variation of garnets, pyroxene andagnetite


Garnet has the chemical formula X3Y2Z3O12, where X is a diva-ent cation such as Ca, Mg, Mn or Fe in eight-fold, dodecahedraloordination, which is replaced by Zn, Y, Na, Li, Be, B, F, Sc, Cu,

6 0.0174516 0.0187481 0.441764

Ga, Ge, Sr, Nb, Ag, Cd, In, U and REE3+ (Gaspar et al., 2008). TheY is the trivalent cation site which includes Fe, Al or Cr in octa-hedral coordination that is substituted by Ti, V, Fe2+, Zr and Sn.Z is Si in tetrahedral coordination which is incorporated by Al,Ti, and Fe3+ site (e.g., Menzer, 1926; Dziggel et al., 2009; Gasparet al., 2008; Smith et al., 2004). The incorporation of REE in gar-net has been widely discussed in the literature (e.g., Gaspar et al.,2008; Ismail et al., 2014; Smith et al., 2004). The partitioning of REEbetween a hydrothermal fluid and a growing garnet is mainly con-trolled by crystal chemistry, crystal-liquid equilibrium (e.g., P, T,fluid composition, complexation), growth rate and surface adsorp-tion (e.g., Smith et al., 2004). Gaspar et al. (2008), in a study ofgold skarn garnets from Crown Jewel, North America, explained theREE incorporation in terms of ionic radii relative to that of Ca, andshowed that REE incorporation could be correlated with changes inmajor element composition. Garnets from the Crown Jewel, NorthAmerica gold skarn (Gaspar et al., 2008), Ocna de Fier Fe-skarn(Nicolescu et al., 1999) and Isle of Skye (Smith et al., 2004) depositare andradite-rich and show enrichment in LREE and depletion inHREE. In these deposits REE patterns are strongly HREE-enrichedwhen garnets are more grossular-rich. In the Sangan iron skarn,REE patterns (Fig. 8) also show that grossular-rich garnets are ingeneral enriched in HREE and depleted in LREE. A LaN/YbN versusFe3+/(Fe3+ + Al) diagram (Fig. 9a) shows that LREE/HREE fractiona-tion does not increase with increasing Fe-content. The ionic radiusand charge disparity between the host cation and its substituent, ata given site, exert the principal control on substitution mechanisms

2+


(Goldschmidt’s first rule). In this case, REE only incorporate Xcations in the dodecahedral position. Incorporation of (REE)3+ byreplacement of X2+ cations requires coupled substitution to main-tain charge balance. There are four mechanisms to maintain charge




F drite-M Zhai

t n of th

b(iXi

ig. 8. Trace element spider diagram of Sangan garnets are shown on the left. ChoncDonough, 1989) (right). Symbols as in Fig. 4. Results from Ismail et al. (2014) and

he references to color in this figure legend, the reader is referred to the web versio

alance including: (i) substitution of a trivalent cation into the ZSi) site as ‘yttrogarnet’ (YAG)-type substitution (Jaffe, 1951); (ii)


ncorporation of a monovalent cation (essentially Na+/Li+) into the site (Enami et al., 1995); (iii) charge compensation via vacancies

n the dodecahedral site or; (iv) by substitution of divalent cations

normalized REE fractionation patterns of Sangan garnets (Chondrite value Sun andet al. (2014) are shown by red and gray field for comparition. (For interpretation ofis article.)

(Mg2+, Fe2+) into the Y site (as in menzerite; Grew et al., 2010).Garnets from the Sangan deposit contain low Na concentrations,


indicating that they did not undergo Na+–REE3+ coupled substitu-tion (Table 4; Gaspar et al., 2008). The high Al2O3 (4.32–12.94 wt.%)along high REE values, might indicate they underwent REE3+–Ca2+




Fig. 9. (a) LREE/HREE fractionation as a function of Fe3+/(Fe3+ + Al) ratio (in atoms per formula unit) of garnets from the Sangan deposit; (b) Eu anomaly as a function ofFe3+/(Fe3+ + Al) ratio; (c) Total REE concentrations versus total Al (in atoms per formula unit) for garnets from the Sangan deposit; (d)–(i) The variation of trace-elementcomposition of garnets from the Sangan deposit with increasing And and/or decreasing Gro components from rim to core; (in atoms per formula unit). Symbols as in Fig. 4.

Sanga

acd

pgne

Fig. 10. Chondrite-normalized REE fractionation patterns of the

nd Al+3- Si+4 coupled substitutions. Other mechanisms such asharge compensation via vacancies in the dodecahedral site areifficult to evaluate but cannot be completely ruled out.

Large ion lithophile elements (LILE) have larger ionic sizes com-


ared with the octahedral and eight coordination sites within thearnet structure (Gaspar et al., 2008), and so, as expected, all gar-ets from the Sangan skarns are extremely LILE-depleted (Fig. 8a, c,, g, i). In comparison, variations in some HFSE in correlation with

n pyroxene (Chondrite values after Sun and McDonough, 1989).

Al suggest that substitution of these elements for Al3+ can occur inthe garnet lattice (Fig. 9d–i).

Pyroxene has the chemical formula XY(Si,Al)2O6 where X isdivalent cations (Ca, Mg, Mn, or Fe) in eightfold coordination, Y


is trivalent cations (Fe, Al, and Cr) in octahedral coordination. Traceelements such as REE3+ may incorporate into pyroxene by replace-ment of Ca2+ in the X-site (Ismail et al., 2014). To achieve chargebalance, coupled substitution involving vacancies (e.g. van Orman




Table 5Selected LA-ICP-MS analyses of pyroxene from Sangan skarn.

A’-019 A’-020 A’-021 A’-022 A’-023 A’-024 A’-025 A’-026 A’-027 A’-028

Si 242628.1 241743.2 241193.9 234611 246872.6 241696 240410.3 306867.9 242883.7 241961.7Ca 168067 168067 168067 168067 168067 168067 168067 168067 168067 168067Sc 2.151687 3.065273 1.827171 2.693725 2.278672 2.64199 2.305715 5.984748 3.006484 3.703724Ti 27.79557 40.39999 78.01337 32.39289 51.39358 44.92676 52.60464 1056.573 37.4958 52.55761Fe3+ 126021.8 106572.8 147719.4 108981.3 126799.2 115584.1 129143.1 270707.5 111833.9 112709.9Fe2+ 124287.4 105139.1 145379.1 107506.6 124592.0 113992.3 127069.8 266507.3 110388.9 111789.9Cu 0.531455 0.497357 0.189301 0.158731 3.809545 0.424459 0.365669 2.037636 0.23986 0.277485Rb 0.164257 0.325693 0.008701 1.078195 0.263376 1.640221 0.475017 18.65972 1.216938 0.14568Sr 4.291617 4.185796 4.632595 4.7149 4.632595 6.419789 4.185796 18.89487 4.891268 3.703724Y 0.035038 0.009524 0.01458 0.005173 0.028924 0.026925 0.020811 1.9083 0.01846 0.007172Zr 0.75603 1.333341 1.164028 1.208707 1.782491 1.146391 0.791304 14.32107 0.91241 0.94298Nb 0.010229 0.013169 0.005056 0.005526 0.026925 0.00776 0.008466 2.627881 0.00776 0.007525Mo 1.071141 0.961793 1.061734 0.880663 1.002945 0.911234 0.941804 1.073492 0.937101 0.932398Ba 1.415646 0.959441 0.023516 1.836577 3.786029 3.268684 0.827753 92.67541 0.993539 0.739569La 0.33745 0.063492 0.009524 0.018577 1.630814 0.014697 0.010347 0.200119 0.024574 0.053498Ce 0.212935 0.145562 0.034098 0.087008 0.594947 0.072899 0.04127 0.673725 0.06749 0.050794Pr 0.017637 0.019636 0.005291 0.016226 0.038095 0.009289 0.008172 0.137449 0.007525 0.004938Nd 0.071135 0.106526 0.025397 0.090418 0.118754 0.100059 0.030688 0.895949 0.048678 0.022928Sm 0.04515 0.029983 0.027866 0.02575 0.036449 0.022458 0.024221 0.380955 0.02234 0.0194Eu 0.017872 0.007172 0.016226 0.00776 0.007172 0.011523 0.01458 0.363318 0.005761 0.008818Gd 0.023045 0.032217 0.021047 0.02234 0.029395 0.023868 0.019871 0.41858 0.023281 0.022105Tb 0.003222 0.003057 0.003269 0.002963 0.003292 0.003763 0.003081 0.0816 0.003527 0.006114Dy 0.01164 0.014109 0.013404 0.01164 0.013639 0.01552 0.014815 0.54674 0.01458 0.016579Ho 0.005173 0.003304 0.003645 0.004115 0.003292 0.003292 0.003998 0.087361 0.00341 0.004938Er 0.009171 0.009289 0.011875 0.011523 0.011052 0.011288 0.008936 0.185774 0.008348 0.01164Tm 0.003175 0.003327 0.003645 0.002928 0.002998 0.004468 0.004233 0.012111 0.002975 0.005409Yb 0.022458 0.01799 0.021399 0.012111 0.012111 0.014227 0.014697 0.084657 0.012816 0.018107Lu 0.003763 0.005409 0.003586 0.002939 0.004703 0.004115 0.003763 0.010582 0.003292 0.005409Hf 0.03786 0.053851 0.034451 0.076661 0.065256 0.033392 0.040565 0.659616 0.057496 0.037155Ta 0.005879 0.002798 0.003034 0.00281 0.003763 0.003527 0.002857 0.05244 0.003645 0.003645W 0.028101 0.033863 0.017637 0.029747 0.026338 0.423283 0.092534 0.09759 0.021634 0.040447204Pb 2.257508 0.858323 0.529103 0.764261 7.419207 3.315715 2.116414 2.480907 1.528521 1.011176206Pb 2.727822 0.338626 0.020106 0.92064 7.901279 4.279859 1.590838 1.994132 1.955331 0.470314207Pb 2.527939 0.343329 0.024927 0.805413 7.395691 3.927124 1.473259 1.769557 1.895366 0.464435208Pb 2.645517 0.312759 0.011875 0.872433 7.607332 4.091734 1.519115 1.96121 1.889487 0.429162Th 0.043269 0.037037 0.003763 0.036332 0.076779 0.030923 0.016696 0.088184 0.022458 0.013404U 0.080424 0.043974 0.003998 0.010347 0.04856 0.008936 0.007878 0.16414 0.025044 0.018342

F ChonC

ecftoRSdab2egt

ig. 11. (a) Primitive mantle-normalized REE patterns of the Sangan magnetite; (b)hondrite values after Sun and McDonough (1989). Symbols as in Fig. 8.

t al., 2001), as well as coupled substitution involving monovalentations (e.g. Na+) has been proposed (Ismail et al., 2014). Pyroxenerom the Sangan deposit contains low Na concentrations, indicatinghat they have not incorporated Na+ into the X site. The presencef Al2O3 might also suggest charge balance via compensation forEE3+ incorporation into the Ca site by Al3+ incorporation into thei site similar to pyroxene from Hillside iron oxide-copper-goldeposit, Yorke Peninsula, South Australia (Ismail et al., 2014). Tottain charge equilibrium, coupled replacements involving Al haseen proposed by Al3+ incorporation into the Si site (Ismail et al.,014). The trace element data on pyroxene also suggest that pyrox-


ne rims possibly grew from a relatively HREE-depleted fluid. Ineneral, the occurrences of HREE-depleted rims in pyroxene crys-als (Fig. 10) also can be attributed to the crystallization of garnets or

drite-normalized multi-elemental patterns of magnetite from the Sangan Fe skarn.

the formation of minerals such as zircon that fractionate HREE fromLREE. Two mechanisms could also be involved: changes in fluidchemical composition (e.g., Dziggel et al., 2009) or simultaneousgrowth of other REE-bearing phases.

Trace elements and REE3+ are also incorporated into magnetite.Chondrite-normalized REE patterns for magnetite from the San-gan skarn are characterized by slight enrichment in LREEs relativeto HREEs (Fig. 11a). However these values are significantly lessthan those from garnet and pyroxene cores. Huang et al. (2015)suggested that there are two generations of massive magnetitesat Bayan Obo, northern China (Huang et al., 2015) including sed-


imentary and hydrothermal magnetite. Magnetite grains fromsedimentary ores have the highest REE contents and show slight tomoderate REE enrichment, whereas those from hydrothermal mag-




Table 6Selected LA-ICP-MS analyses of magnetite from Sangan skarn.

Sample (ppm) CNF-04 CNF-05 CNF-06 B3-07 B3-08 B3-09 Cs31-010 Cs31-011 A3-022 A3-023

Si 1363.96 2065.614 1717.343 1483.879 1437.175 1328.42 1051.798 1400.187 5207.304 2172.098Ca 66.18346 121.5717 342.1613 58.01118 21.71526 51.60334 68.14866 117.1743 529.2728 115.2473Sc 0.045189 0.067784 0.061821 0.124263 0.017196 0.041529 0.067685 0.052147 0.75127 0.473489Ti 180.6818 178.3344 165.8729 2.968693 5.476670 11.49192 51.17734 48.29875 35.45489 13.28611Fe3+ 243931.5 243931.5 243931.4 252196.5 252196.5 252196.5 204717.6 204717.6 245365.6 245365.6Fe2+ 221124.3 221129.6 221526.9 231355.1 231258.0 231097.4 187136.1 187597.5 243344.3 244398.7Cu 0.077512 0.150631 0.082847 0.036338 0.055156 0.079814 0.118778 0.165394 63.87688 26.49647Rb 0.648341 0.533171 0.744368 0.276429 0.158330 0.196615 1.722416 1.677643 0.520838 0.179295Sr 1.531414 2.049208 6.750148 0.605419 0.598281 0.815012 1.917307 2.794316 3.342835 1.275265Y 0.027616 0.018044 0.022218 0.05149 0.008663 0.005937 0.009297 0.008875 1.167941 0.684981Zr 0.026957 0.083161 0.235047 0.156059 0.026280 0.021965 0.061101 0.127469 0.161933 0.014205Nb 0.045095 0.116739 0.241009 0.014373 0.025274 0.049381 0.003476 0.010482 0.048927 0.005556Mo 0.049269 0.082847 0.103559 0.043476 0.032445 0.046072 0.057941 0.053200 3.560640 1.291048Ba 2.535620 3.113039 6.52734 1.275078 0.899368 0.947386 3.510673 8.514632 28.82288 4.842218La 0.054447 0.113287 0.320718 0.044157 0.030530 0.043411 0.125889 0.112194 0.435610 1.991812Ce 0.099793 0.104186 0.275215 0.064111 0.037344 0.056292 0.181986 0.196471 0.580814 2.992453Pr 0.009948 0.008536 0.02294 0.008273 0.002174 0.005159 0.018778 0.022439 0.074180 0.328286Nd 0.046758 0.034833 0.065587 0.048343 0.017390 0.021024 0.069792 0.086648 0.318816 1.376276Sm 0.016318 0.011925 0.009414 0.013951 0.007592 0.008144 0.009481 0.013168 0.111112 0.321973Eu 0.003483 0.00317 0.002887 0.003082 0.003147 0.003082 0.001870 0.002344 0.059028 0.122476Gd 0.016946 0.011611 0.012553 0.012329 0.007592 0.005743 0.008691 0.010271 0.171403 0.266732Tb 0.002354 0.001444 0.001789 0.003439 0.001687 0.002823 0.001106 0.001370 0.029988 0.037248Dy 0.007343 0.00728 0.006904 0.016449 0.006262 0.005872 0.005926 0.005320 0.208020 0.160355Ho 0.002259 0.001883 0.002008 0.001947 0.001622 0.001200 0.001528 0.002002 0.048927 0.024180Er 0.00477 0.008191 0.004864 0.007495 0.003796 0.003666 0.003424 0.003608 0.115216 0.043877Tm 0.001349 0.002354 0.001506 0.001428 0.001752 0.001395 0.001264 0.001238 0.013131 0.006692Yb 0.008253 0.009414 0.010419 0.006619 0.007981 0.006976 0.005320 0.004346 0.079231 0.041036Lu 0.00204 0.002416 0.001695 0.001914 0.001784 0.001038 0.001185 0.001422 0.008807 0.005082Hf 0.004174 0.006025 0.005649 0.005321 0.005159 0.004477 0.009587 0.005847 0.007955 0.009628Ta 0.009163 0.019519 0.033296 0.002109 0.001298 0.001525 0.002370 0.001817 0.003472 0.002904W 0.057114 0.063704 0.071863 0.160277 0.672904 2.209486 0.031867 0.042402 6.234277 0.344069204Pb 1.176804 1.032449 1.045002 1.031742 0.425026 0.279025 0.526733 0.695287 0.830185 0.688138206Pb 0.785477 1.248981 0.828784 1.301034 0.295896 0.318283 0.729525 0.658679 1.262638 1.010111207Pb 0.711417 1.154837 0.803365 1.135566 0.259558 0.334505 0.647618 0.622071 1.208976 0.918569208Pb 0.677525 1.208185 0.810896 1.262100 0.272211 0.320878 0.589150 0.627602 1.234229 0.972232Th 0.023285 0.006057 0.020618 0.050776 0.003666 0.009604 0.006610 0.009165 0.051768 0.147097U 0.015879 0.026863 0.009822 0.018266 0.010123 0.012589 0.057045 0.059916 1.619334 0.467176

F erals

s body

nfn

ig. 12. Results of the baro-acoustic decrepitation of fluid inclusions hosted by mintage) from B and C orebodies and garnet (A-1) (silicate dominant stage) from A ore


etite show light REE enrichment. The REE patterns of magnetiterom Sangan are quite similar to those from hydrothermal mag-etite at Bayan Obo, northern China (Huang et al., 2015), indicating

comprising high grade massive magnetite ores (B-4 and C-4) (magnetite dominant.


a similar origin. The mineral chemistry of magnetite deposited fromhydrothermal fluids is mostly controlled by: (1) fluid composition,(2) temperature, (3) pressure, (4) cooling rate, (5) oxygen or sul-


ING ModelC

e der E

ftteaavsiaclAnwsCatstSraama5d2nwatt

6

ef1+inwFmirgchacAIaflsa2atlE



ur fugacity, (6) silica activity (Nadoll et al., 2014; and referencesherein) and (7) composition of host rocks (Nadoll, 2010). Several ofhem have evidently controlled the magnetite formation in differ-nt portions of the Sangan deposits. Low abundances of Al2O3, TiO2nd MgO in magnetite from Sangan indicate exsolution of ilmenitend spinel could not have occurred. The MgO content of magnetitearies as follow: 0.01–0.26 wt.% in sample A-3; up to 0.03 wt.% inample B-2, and 0.02–0.26 in sample Cn-1. The relatively high MgOn A and Cn orebodies is consistent with their formation at rel-tively high temperatures (Nadoll et al., 2012). Magnetites alsoontain moderate to high SiO2 (0.06–3.43 wt.% for A, B, and Cn),ow MnO (0.01–0.12 wt.%, for A, B, and Cn) and low to moderatel2O3 (0.05–0.43 wt.% for A, B, and Cn) contents (Table 3). Mag-etite contains submicrometer-to-micrometer inclusions of quartzhich indicates the oxide precipitate was chemically impure and

hows high Si content in some samples in Sangan. Magnetite in thes orebody has low MgO (0.03–0.07 wt.%), low SiO2 (<0.024 wt.%)nd MnO (0.09–0.13) which is consistent with calcic skarns. In con-rast, the range of the Cu contents of magnetite from A orebody isignificantly higher than those from B, Cn and Cs. This is attributedo difference of coexisting sulfides content in all orebodies of theangan deposit. Titanium and V are a powerful tool to monitor foreduction-oxidation conditions and the influence of mafic materi-ls on magnetite formation. Ti has only Ti(IV) in hydrothermal fluidsnd hence the partition coefficient is rather consistent betweenagnetite and fluids. The Ti/V ratio of magnetite from the western

nd eastern parts of the Tieshan deposit, China, ranges from 1.32 to.24, and 1.31 to 10.34, respectively, indicating a relatively reducedepositional environment in the western ore body (Wang et al.,017). The Ti/V (2.00–26.0 with 12/4 +2 as an average) ratio of mag-etite from the Cn orebody is higher than those from A (2.0–11.0ith 5.7 +2 as an average), B (2.0–11.0 with 2.9 +2 as an average)

nd Cs (0.25–5.00 with 1.7 +0.2 as an average), that indicate a rela-ively oxidized depositional environment for Cn orebody comapredo the other orebodies in the Sangan region.

.2. Implications for skarn hydrothermal fluid evolution

Previous studies show that �18O values for water in isotopicquilibrium with garnet and magnetite from the Sangan area rangerom +7.7 to +11.6‰ and +9.8 to +14.2‰, respectively (Mazaheri,996), which are higher than those of magmatic water (+5.5 to10‰: Hoefs, 2009). (Rose et al. (1985)) measured the oxygensotope composition of magnetite from southeastern Pennsylva-ia (�18O = +5.2 to +10.0‰) and suggested that 18O enrichment ofater in equilibrium with magnetite (�18OH2O = +12.3 to +17.4‰;

ig. 10a) was the result of a high temperature reaction betweenagmatic fluid and carbonate country rocks enriched in 18O. It

s thus likely that the �18O compositions of hydrothermal fluidsesponsible for crystallization of garnet and magnetite in the San-an skarn increased due to interaction of magmatic water witharbonate country rocks. The minerals from the calc-silicate stageave major roles in governing REE fractionation patterns at Sangannd can thus map the evolution of the mineralizing system. Garnetompositions change from And35 to And65 in the core and And50 tond70 in the rim which suggests reducing conditions. Based on LA-

CP-MS analyses of zoned garnets, HREE-rich patterns of the corend rim indicate that garnet crystallized from a relatively HREE-richuid. Previous studies show that LREE chloride complexes are moretable at higher temperature whereas HREE chloride complexesre more stable at low temperatures (Williams-Jones and Heinrich,005). This can change partitioning of the REE between the fluid


nd garnet. Bau (1991) suggested that The Eu3+/Eu2+ redox poten-ial in hydrothermal fluids primarily depends on temperature andess on pH and pressure. However, at temperatures above 250 ◦C,u2+ should predominate in hydrothermal solutions (Sverjensky,

PRESSrde xxx (2017) xxx–xxx 19

1984). Therefore, at relatively high temperatures observed in skarnsystems, Eu can fractionate from the other REEs (Bau, 1991). Innearly neutral conditions, the REE patterns of the fluid are relativelyHREE-enriched and LREE-depleted (e.g., Bau, 1991). If fluid-rockinteraction happens under mildly acidic conditions, the REE patternof the fluid would be controlled by sorption processes and provideEu as Eu 2+ to a positive Eu anomaly in the fluid. The REE patterns inthe zoned garnets in the Sangan skarn deposit are consistent withthose patterns mentioned above (Fig. 8), suggesting they formed innear neutral to mildly acidic conditions.

Mayanovic et al. (2007) show that in hydrothermal fluids, Cl−isa carrier for Eu2+ and a poor transporter for REE3+. The local positiveEu anomalies in garnet and pyroxene cores might be explained bythe presence of ligands, such as Cl−, which transfer Eu2+. The exis-tence of a Eu anomaly suggests that Eu is present in the divalentstate as opposed to the trivalent state of the other REEs, and thushigh temperature and reducing conditions can be inferred (Gasparet al., 2008), which is consistent with microprobe results. In con-trast, chondrite-normalized REE patterns for pyroxene rims andmagnetite are characterized by LREE enrichment relative to HREEswithout Eu anomalies (Figs. 10 and 11). These characteristics indi-cate that pyroxene rims and magnetite possibly originated fromfluids which were depleted in HREE by crystallization of garnets orother minerals (Zhai et al., 2014). Europium is present as Eu3+ underoxidizing conditions (Gaspar et al., 2008), and this likely explainsits similar behavior to other REEs in magnetite and pyroxene rims(Figs. 10 and 11).

Similar to garnet from Sangan, trends described by Zhai et al.(2014) for garnets from Zhongdian area, NW Yunnan Province,China, show that they have high HREE, Th, U, Pb and Y and low inLREE, Sr, Ba, La, and Ti (Fig. 8b). Ismail et al. (2014) suggest thatREE distribution patterns in mineral assemblages, garnet, diop-side and feldspar define the evolution from early skarn throughthe calc-silicate dominant stage to the magnetite stage. Theyalso concluded that distinct REE patterns from different stages ofhydrothermal fluid evolution are related to redox changes through-out different lithologies in the Hillside IOCG-skarn. However, Zhaiet al. (2014) also show that garnet cores were formed from afluid that was generally HREE-rich, with relatively high �REE,high HREE/LREE ratios and negative Eu anomalies; whereas rimscrystallized from a fluid that was typically LREE-rich, with rela-tively low �REE, low HREE/LREE ratios and positive Eu anomalies.They proposed that the hydrothermal fluid evolved from nearneutral pH and oxidizing conditions to acidic and reducing condi-tions. The chondrite-normalised REE fractionation patterns for theskarn zone at Ali Abad skarn depicted by Zamanian and Radmard(2016) are similar to those from the Sangan skarn. Zamanian andRadmard (2016) showed the highest and lowest REE-enriched pat-terns for skarn and ore zones, respectively. They concluded thathydrothermal fluids responsible for garnet formation were mostlyof magmatic origin with high REE concenteration and for mag-netite formation were of magmatic and meteoric water origin withlow REE concentrations. The chondrite-normalised REE fractiona-tion patterns for garnet depicted by Xu et al. (2016) are differentfrom the Sangan skarns, however, garnet from Sangan shows asimilar YAG-type substitution mechanism. These results confirmthat variations in fluid composition and physicochemical condi-tions are considered as major controls on incorporation of traceelements, including REEs, into garnet rather than YAG-type sub-stitution mechanisms. As U solubility decreases with decreasingfO2 of the hydrothermal fluid, this would increase U incorporationinto garnet (e.g., Smith et al., 2004). In contrast, increasing fluid


fO2 would increase U solubility and reduce U incorporation intomagnetite (Fig. 13). A similar process was involved to describe Uvariations in garnets from the Xieertala Fe-Zn skarn deposit, north-ern China (Zhai et al., 2014).


ARTICLE ING ModelCHEMER-25446; No. of Pages 21

20 F. Sepidbar et al. / Chemie der E

F�

ahrptw

7

(

(

(

(

A

Csorif

ig. 13. Covariation of REE3+ with U in the Sangan garnet. Note the variations ofREE with U in the Sangan magnetite have been shown by grey field.

Based on the baro-acoustic decrepitation results, REE patternsnd Eu anomalies in zoned garnets, pyroxene and magnetite, theydrothermal fluid in the skarn evolved from a near neutral pH,educed fluid with low fO2, which was relatively HREE-rich, withositive Eu2+ and U-rich in the silicate-dominant stage, progressingo an acidic and oxidized fluid which was relatively HREE-depleted,ith no Eu2+ anomaly and U-poor in the magnetite-dominant stage.

. Conclusions

1) In situ LA-ICP-MS analyses of zoned garnets, pyroxene andmagnetite from Sangan skarn show that garnets were formedfrom a fluid that was generally HREE-rich, with relatively high�REE, high HREE/LREE ratios, positive Eu anomalies and U-rich characteristics whereas pyroxene rims and magnetite werecrystallized from a fluid that was typically LREE-rich, withrel-atively low �REE, low HREE/LREE ratios, with no Eu anomaliesand U-poor characteristics.

2) Analysis of REE patterns, Eu anomalies and U contents in zonedgarnet, pyroxene and magnetite shows that the hydrothermalfluid in the skarn changed from a near neutral to mildly acidicpH and low fO2 in the silicate stage towards acidic conditionsand high fO2 in the magnetite stage.

3) The dominant substitution mechanism for REE distributionwhitin the garnet is YAG-type. The results show that variationsin fluid composition and physicochemical conditions ratherthan YAG-type substitution mechanisms are considered to havea major control on incorporating trace elements and REEs intothe garnet.

4) The calc-silicate dominant stage was progressively formedat temperatures around 350–450 ◦C as indicated by mineralchemistry and the decrepitation analysis of garnet. The mag-netite dominant stage was progressively formed at similaror slightly higher temperature (around 400 ◦C) as indicatedby the decrepitation analysis of magnetite. The Lack of anylow-temperature peak around 300 ◦C indicates an absence ofCO2-rich fluid inclusions.

cknowledgements

The authors would like to thank Professor Changqian Ma fromUG for his assistance during analyses at the GPMR. Constructiveuggestions by Prof. Nigel J Cook, Dr. Cristiana Ciobanu (University


f Adelaide, Australia) and Prof. Jian-Feng Gao (State Key Labo-atory of Ore Deposit, China Academy of Science) substantiallymproved the manuscript. We would like to thank Dr. Tomas Magnaor his comments and editorial handling of the paper and Dr.

PRESSrde xxx (2017) xxx–xxx

Jaayke Knipping and two anonymous reviewer for reviewing themanuscript and providing us with invaluable comments. This studywas performed as part of a collaborative multidisciplinary researchproject on the Sabzevar-Doruneh magmatic belt chiefly involvingthe Sangan, funded by the School of Earth Resources and State KeyLaboratory of Geological Processes and Mineral Resources, ChinaUniversity of Geosciences, Wuhan. We gratefully acknowledge thefinancial support from the geochemical analyses funded by Min-istry of Science and Technology of China (2012CB416802).

References

Alavi Naini, M., 1982. Geological Quadrangle Map of Khaf: Series Sheet 8059,Ministry of Mines and Metals. Geological survey of Iran, Tehran.

Armbruster, T., Birrer, J., Libowitzky, E., Beran, A., 1998. Crystal chemistry ofT-bearing andradites. Eur. J. Mineral. 5, 907–921.

Bau, M., 1991. Rare-earth element mobility during hydrothermal andmetamorphic fluid/rock interaction and the significance of the oxidation stateof Europium. Chem. Geol. 93, 219–230.

Burlinson, K., 1984. Exploration for gold at Pine Creek and Tennant Creek, N.T andat Halls Creek, W.A, using the fluid inclusion decrepitation technique. In: TheAus. I.M.M. Conference, Darwin, pp. 373–375.

Burlinson, K., 1988. An instrument for fluid inclusion decrepitometry andexamples of its application. Bull. Miner. Res. Explor. 111, 3–4, 267–278.

Burlinson, K., 1990. The BGS Model 105 Decrepitometer. http://www.appliedminex.com/decrep/general/model105.htm.

Burlinson, K., 2001. Decrepitation in Haematite and Magnetite. http://appliedminex.com/decrep/general/haemag.htm.

Burlinson, K., 2012. Fluid inclusion studies on opaque minerals. ACROFI-4Conference http://appliedminex.com/decrep/orals/bne12/opaquedecrep.htm.

Canil, D., Grondahl, C., Lacourse, T., Pisiak, L.K., 2016. Trace elements inmagnetitefrom porphyry Cu-Mo-Au deposits in British Columbia, Canada. OreGeol. Rev. 72, 1116–1128.

Chakraborty, S., Ganguly, J., 1991. Compositional zoning and cation diffusion ingarnets. In: Diffusion, Atomic Ordering, and Mass Transport. Springer, US, pp.120–175.

Chang, Z.S., Meinert, L.D., 2004. The magmatic-hydrothermal transition – evidencefrom quartz phenocryst textures and endoskarn abundance in Cu-Zn skarns atthe Empire Mine Idaho, USA. Chem. Geol. 210, 149–171.

Cheng, H., Zhang, C., Vervoort, J.D., Lu, H.H., Wang, C., Cao, D.D., 2012. Zircon U-Pband garnet Lu-Hf geochronology of eclogites from the Lhasa Block, Tibet. Lithos155, 341–359.

Ciobanu, C.L., Cook, N.J., 2004. Skarn textures and a case study: the Ocna deFier-Dognecea orefield Banat, Romania. Ore Geol. Rev. 24, 315–370.

Dare, S.A.S., Barnes, S.J., Beaudoin, G., Meric, J., Boutroy, E., Potvin-Doucet, C., 2014.Trace elements in magnetite as petrogenetic indicators. Miner. Depos. 49,785–796.

Dupuis, C., Beaudoin, G., 2011. Discriminant diagrams for iron oxide trace elementfingerprinting of mineral deposit types. Miner. Depos. 46, 319–335.

Dziggel, A., Wulff, K., Kolb, J., Meyer, F.M., Lahaye, Y., 2009. Significance ofbell-shaped growth zoning in hydrothermal garnet: evidence from theNavachab gold deposit, Namibia. Chem. Geol. 262, 262–276.

Enami, M., Cong, B., Yoshida, H., Kawabe, I., 1995. A mechanism for Naincorporation in garnet: an example from garnet in orthogneiss from the Su-Luterrane, eastern China. Am. Mineral. 80, 475–482.

Fernando, G., Hauzenberger, C.A., Baumgartner, L.P., Hofmeister, W., 2003.Modeling of magnetite dominant diffusion zoning in garnet: evidence for slowcooling of granulites from the Highland Complex of Sri Lanka. Mineral. Petrol.78, 53–71.

Frietsch, R., Perdahl, J.A., 1995. Rare earth elements in apatite and magnetite inKiruna type iron ores and some other iron ore types. Ore Geol. Rev. 9, 489–510.

Gaspar, M., Knaack, C., Meinert, L.D., Moretti, R., 2008. REE in skarn systems: aLA-ICP-MS study of garnets from the Crown Jewel gold deposit. Geochim.Cosmochim. Acta 72, 185–205.

Golmohammadi, A., Karimpour, M.H., Malekzadeh, A., 2015. Alterationmineralization, and radiometric ages of the source pluton at the Sangan ironskarn deposit, northeastern Iran. Ore Geol. Rev. 65, 545–563.

Grew, E.S., Marsh, J.H., Yates, M.G., Lazic, B., Armbruster, T., Locock, A., Bell, S.W.,Dyar, M.D., Bernhardt, H.-J., Medenbach, O., 2010. Menzerite-(Y), a newspecies, {(Y, REE) (Ca, Fe2+)2}[(Mg, Fe2+)(Fe3+, Al)](Si)3 O12, from a felsicgranulite, Parry Sound, Ontario, and a new garnet end-member,{Y2Ca}[Mg2](Si3)O12. Can. Mineral. 48, 1171–1193.

Hariya, Y., Kimura, M., 1978. Optical anomaly garnet and its stabiliry field at highpressures and temperatures. J. Fac. Sci. 18, 611–624, Hokkaido University,Series IV.

Hoefs, J., 2009. Stable Isotope Geochemistry. Springer-Verlag, Berlin, Heidelberg,


New York.Hu, H., Li, J.-W., Lentz, D., Ren, Z., Zhao, X.-F., Deng, X.-D., Hall, D., 2014.

Dissolution-reprecipitation process of magnetite from the Chengchaoirondeposit: insights into ore genesis and implication for in-situ chemicalanalysisof magnetite. Ore Geol. Rev. 57, 393–405.


http://refhub.elsevier.com/S0009-2819(17)30077-6/sbref0005











































































































http://www.appliedminex.com/decrep/general/model105.htm








http://appliedminex.com/decrep/general/haemag.htm







http://appliedminex.com/decrep/orals/bne12/opaquedecrep.htm













































































































































































































































































































































































































ING ModelC

e der E

H

I

J

J

K

K

KK

M

M

MM

M

M

M

M

N

N

N

N

of oscillatory zoned garnets from the Xieertala Fe-Zn skarn deposit, northern



uang, X., Zhou, M.-F., Qiu, Y.Z., Qi, L., 2015. In-situ LA-ICP-MS trace elementalanalyses of magnetite: the Bayan Obo Fe-REE-Nb deposit, North China. OreGeol. Rev. 65, 884–899.

smail, R., Ciobanu, C.L., Cook, N.J., Schmidt Mumm, A., Wade, B., Giles, D., Teale,G.S., 2014. Rare Earths and other trace elements in minerals from skarnassemblages Hillside iron oxide-copper-gold deposit, Yorke Peninsula, SouthAustralia. Lithos 184–187, 456–477.

affe, H.W., 1951. The role of yttrium and other minor elements in the garnetgroup. Am. Mineral. 36, 133–155.

amtveit, B., Wogelius, R.A., Fraser, D.G., 1993. Zonation patterns of skarn garnets:records of hydrothermal system evolution. Geology 21, 113–116.

arimpour, M.H., 2004. Mineralogy, alteration, source rock, and tectonic setting ofIron- Oxides Cu-Au deposits and examples of Iran. In: 11th IranianCrystallography and Mineralogy Society of Iran Conference, University ofYazad, pp. 184–189.

ermani, A., Forster, H., 1991. Petrography, Mineralogical and geochemicalinvestigations of the Sangan Iron ore deposit, northeastern Iran. Proceedings ofThird Mining Symposium of Iran.

im, H.S., 2006. Deformation-induced garnet zoning. Gondwana Res. 10, 379–388.nipping, J.L., Bilenker, L.D., Simon, A.C., Reich, M., Barra, F., Deditius, A.P., Wälle,

M., Heinrich, C.A., Holtz, F., Munizaga, R., 2015. Trace elements inmagnetitefrom massive iron oxide-apatite deposits indicate a combinedformation byigneous and magmatic-hydrothermal processes. Geochim.Cosmochim. Acta 171, 15–38.

artin, L.A.J., Ballèvre, M., Boulvais, P., Halfpenny, A., Vanderhaeghe, O., Duchêne,S., Deloule, E., 2011. Garnet re-equilibration by coupleddissolution-reprecipitation: evidence from textural, major element andoxygen isotope zoning of ‘cloudy’ garnet. J. Metamorph. Geol. 29, 213–231.

azaheri, S.A., 1996. Petrological Studies of Skarns from Marulan, New SouthWales, Australia and Sangan, Khorassan, Iran. University of California, Ph. Dthesis, 318p.

einert, L.D., 1992. Skarns and skarn deposits. Geosci. Can. 19, 145–162.einert, L.D., 1997. Application of skarn deposit zonation models to mineral

exploration. Explor. Min. Geol. 6, 185–208.einert, L.D., Nicolescu, S., 2005. World skarn deposits. In: Economic Geology

100th Anniversary Volume., pp. 299–336.einert, L.D., Dipple, G.M., Nicolecu, S., 2005. World skarn deposits. In: Economic

Geology 100th Anniversary Volume., pp. 299–336.enzer, G., 1926. Die kristallstructure von granat. Zeitschrift für Kristallographie

63, 157–158.irnejad, H., Blourian, G.H., Kheikhah, M., Akrami, M.A., Tutti, F., 2008. The

chemistry of hydrothermal magnetite: a review. Ore Geol. Rev. 61, 1–32.adoll, P., Mauk, J.L., Hayes, T.S., Koenig, A.E., Box, S.E., 2012. Geochemistry of

magnetite from hydrothermal ore deposits and host rocks of theMesoproterozoic Belt Supergroup United States. Econ. Geol. 107, 1275–1292.

adoll, P., Angerer, T., Mauk, J.L., French, D., Walshe, J., 2014. The chemistry ofhydrothermal magnetite: a review. Ore Geol. Rev. 61, 1–32.

adoll, P., 2010. Geochemistry of Magnetite from Hydrothermal Ore Deposits andHost Rocks- Case Studies from the Proterozoic Belt Supergroup,


Cu-Mo-porphyry+ Skam and Climax-Mo Deposits in the Western UnitedStates. The Uiversity of Auckland, Auckland.

icolescu, S.Ë., Cornell, D.H., Bojar, A.V., 1999. Age and tectonic setting of BocsËaand Ocna de Fier-Dognecea granodiorites (southwest Romania) and ofassociated skarn mineralisation. Miner. Depos. 34, 743–753.

PRESSrde xxx (2017) xxx–xxx 21

Ravna, E.K., 2000. The garnet-clinopyroxene Fe2+-Mg geothermometer: anupdated calibration. J. Metamorph. Geol. 18, 211–219.

Rose, A.W., Herrick, D.C., Deines, P., 1985. An oxygen and sulfur isotope study ofskarn-type magnetite deposits of the Cornwall type, southeasternPennsylvania. Econ. Geol. 80, 418–443.

Russell, J.K., Dipple, G.M., Lang, J.R., Lueck, B., 1999. Major element discriminationof titanian andradites from magmatic and hydrothermal environments: anexample from the Canadian Cordillera. Eur. J. Mineral. 6, 919–935.

Samadi, R., Mirnejad, H., Kawabata, H., Valizadeh, M.V., Harris, C., Gazel, E., 2014.Magmatic garnet in the Triassic (215 Ma) Dehnow pluton of NE Iran and itspetrogenetic significance. Int. Geol. Rev. 5, 596–621.

Schmidt, A., Mezger, K., O’Brien, P.J., 2011. The time of eclogite formation in theultrahigh pressure rocks of the Sulu terrane: constraints from Lu-Hf garnetgeochronology. Lithos 125, 743–756.

Smith, M.P., Henderson, P., Jeffries, T.E.R., Long, J., Williams, C.T., 2004. The rareearth elements and uranium in garnets from the Beinn an Dubhaich Aureole,Skye, Scotland, UK: constraints on processes in a dynamic hydrothermalsystem. J. Petrol. 45, 457–484.

Sosnicka, M., Bakker, J.R., Broman, C., Pitcairn, I., Paranko, I., Burlinson, K., 2015.Fluid types and their genetic meaning for the BIF-hosted iron ores Krivoy Rog,Ukraine. Ore Geol. Rev. 68, 171–194.

Somarin, A.K., 2004. Garnet composition as an indicator of Cu mineralization:evidence from skarn deposits of NW Iran. J. Geochem. Explor. 81, 47–57.

Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanicbasalts: implication for mantle composition and processes. In: Saunder, A.D.,Norry, M.J. (Eds.), Magmatism in the Ocean Basins. Geological Society, London,pp. 313–345, Special Publications 42.

Sverjensky, D.M., 1984. Europium redox equilibria in aqueous solution. Earth andPlanetary. Sci. Lett. 67, 70–78.

van Orman, J.A., Grove, T.L., Shimizu, N., 2001. Rare earth element diffusion indiopside: influence of temperature, pressure, and ionic radius and an elasticmodel for diffusion in silicates. Contributions to Mineralogy and Petrology 141,687–703.

Wang, M., Wanga, W., Liu, K., Michalak, P.P., Weid, K., Hue, M., 2017. In-situLA-ICP-MS trace elemental analyzes of magnetite: the Tieshan skarn Fe-Cudeposit, Eastern China. Chemie der Erde - Geochem. 77, 169–181.

Whitney, D.L., 1996. Garnets as open systems during regional metamorphism.Geology 24, 147–150.

Williams-Jones, A.E., Heinrich, C.A., 2005. Vapour transport of metals and theformation of magmatic-hydrothermal ore deposits. Econ. Geol. 100,1287–1312.

Xu, J., Ciobanu, C.L., Cook, N.J., Zheng, Y., Sun, X., Wade, B.P., 2016. Skarn formationand trace elements in garnet and associated minerals from Zhibula copperdeposit, Gangdese Belt, southern Tibet. Lithos 262, 213–231.

Zamanian, H., Radmard, K., 2016. Geochemistry of rare earth elements in the BabaAli magnetite skarn deposit, western Iran – a key to determine conditions ofmineralization. Geologos 22 (1), 33–47.

Zhai, D.G., Liu, J.J., Zhang, H.Y., Wang, J.P., Su, L., Yang, X.A., Wu, S.-H., 2014. Origin


China: in situ LA-ICP-MS evidence. Lithos 190–191, 279–291.Zhao, W.W., Zhou, M.F., 2015. In situ LA-ICP-MS trace elemental analyses of

magnetite: the Mesozoic Tengite skarn Fe deposit in the Nanling Rang, SouthChina. Ore Geol. Rev. 65, 872–883.




























































































































































































































































































































































































































































































































































































































































































































































































































G Model ARTICLE IN PRESS - appliedminex.com · skarn deposit Sangan Iran a b s t r a c t The Sangan...

Documents

Transcript of G Model ARTICLE IN PRESS - appliedminex.com · skarn deposit Sangan Iran a b s t r a c t The Sangan...